Wolfram Computation Meets Knowledge

Wolfram Technology Conference 2022, Champaign, IL (Home of Wolfram's Headquarters)

Thank you to all attendees and speakers for making the Wolfram Technology Conference a success!

Presentations

Tuesday, October 18

Stephen Wolfram Opening Keynote

Stephen Wolfram

Founder and CEO, Wolfram Research

Watch Video

Wednesday, October 19

13.2 Front End Overview

John Fultz Download Presentation

Configuring and Managing Wolfram Language Kernels

Roman Maeder Download Presentation Watch Video

From within a Wolfram Language session, it is possible to connect to other Wolfram Language engines (kernels) both locally and on remote computers using a variety of connection methods. Such kernels can be used for ad-hoc evaluations, as session kernels of a notebook or bundled for parallel computation. Support for discovering, configuring and managing kernel resources has been modernized over the last three language versions, and is now available in a single coherent way across a wide variety of environments, from ad-hoc collections of workstations to managed clusters, supercomputers and cloud compute services. We give an overview over the kernel connection methods and their intended use cases. Connection methods discussed include local kernels, remote access (SSH), WSTPServer and Wolfram Lightweight Grid.

Educational Innovation

Paul Abbott Download Presentation Watch Video

For the last six years, the Wolfram Summer School has included an Educational Innovation track. There are many reasons why the school is an excellent place at which to develop and implement educational innovation:

  • Extensive high-quality mentoring from experts and Wolfram Research developers, sharing their experience in making educational technologies
  • Stephen Wolfram is an active participant and provides general guidance to—and review of—all projects
  • Quality feedback from and interaction with—other like-minded educators
  • Intensive, hands-on three-to-four-week format including in-depth training on Wolfram Language technology development
  • Lectures and discussions on computational thinking and current strategies
  • Quickly develop education tools using Wolfram technologies
  • The focus of projects is to change the way people learn and teach—for example, creating tools for students to use or management platforms for teachers

In this presentation I will give an overview of the Wolfram Summer School and highlight some presentations, published on the Wolfram Community forum. So, if you are interested in educational innovation, I hope to convince you to attend a Wolfram Summer School in the not-too-distant future.

Iterators: The Next Iteration

Roman Maeder Download Presentation Watch Video

Iterators are a generalization of lists that are accessed one element at a time. Iterators allow us to work with data whose length is infinite or unknown, and they avoid the explicit generation of all elements at the same time, by using incremental generators. First presented a year ago, this project has now matured, and I would like to share some of the improvements made, focusing on programming techniques for paclet design, seamless integration of iterators into the Wolfram Language and compiled iterators.

Software Development in the Notebook Front End

Ian Hojnicki Download Presentation

Wolfram Quantum Framework

Nikolay Murzin, Mads Bahrami Download Presentation

UX Projects: Assistance in Authoring Notebooks and Writing Wolfram Language Code

Noah Hardwicke

Version 13.1 of the Wolfram Language saw the introduction of a general-purpose notebook toolbar, the first of several planned UX additions to the front end to aid notebook authoring and code writing. In this talk I will review the notebook toolbar and discuss some upcoming features, including a new select and wrap tool, new message reporting and a scoping assistant for managing the variables of Module and related constructs.

Workshop: Machine Learning

Giulio Alessandrini, Jesse Galef Download Presentation Watch Video

Progress in Geo Computation

Jose Martin-Garcia Download Presentation Watch Video

Introducing Astro Computation

Jose Martin-Garcia Download Presentation Watch Video

Green Energy Modeling

Ankit Naik, Jan Brugard Download Presentation Watch Video

In this presentation, you will understand the relevance of modeling in the green energy sector using a wind farm model. You will get to know how some of our industrial customers are using System Modeler to make policy decisions.

Using Resource Functions for K–College Math

Jeremy Stratton-Smith, Jordan Hasler, AnneMarie Torresen, Enrique Zeleny Vazquez, Nicholas Brunk Download Presentation Watch Video

We will discuss the 140+ resource functions that the Wolfram|Alpha Math Content team has published, primarily in the service of Wolfram|Alpha Notebook Edition. These functions build upon and sometimes run in parallel to kernel functions, and often are aligned to questions that students commonly encounter, provide greater description in their results and/or include interactive content to extend the built-in functionalities of the Wolfram Language.

New in Tree Data & Computation

Ian Ford Download Presentation Watch Video

Tree is a new fundamental construct in the Wolfram Language added in Version 12.3 used to represent rooted, ordered, labeled trees. In this talk, we will present a variety of new features added in Version 13.1, including support for associations between keys and children, new ways of constructing tree data such as XML and JSON, options for detailed control of labeling and styling, and new functional programming operations.

Aircraft Design in System Modeler

Niko Era-Esko Download Presentation Watch Video

Developed in collaboration with Saab Aircraft, the new Aircraft library offers a convenient way to model aircraft dynamics to study different designs.

Meet-Up: Engineering Education

Devendra Kapadia

The aim of this meet-up is to discuss the use of Wolfram technology in engineering education. Participants will learn about current projects related to Mathematica and System Modeler in this area, and will also have an opportunity to share their own ideas about learning and teaching engineering concepts and skills in the modern world. All are welcome to attend the meet-up, including professional engineers, college faculty, parents of prospective engineering students and others who are interested in this critical area of higher education.

Workshop: Active Learning Notebooks for College Mathematics

Jordan Hasler Download Presentation Watch Video

Modern learning spaces transcend mechanical pencil-and-paper calculations by combining computation, visualization and concept exploration. Wolfram|Alpha Notebook Edition facilitates all three of these while quietly teaching programming in the Wolfram Language along the way. The breadth and depth of pedagogy tools for college mathematics are demonstrated through standalone student activities. Topics range from calculus, linear algebra and differential equations to probability and statistical and data analysis.

Analyzing Online Sentiment Towards Marvel TV Shows

Rohan Mehta Download Presentation Watch Video

Recently, the Marvel Cinematic Universe has expanded to several TV shows through the Disney+ streaming platform. The episodic format of these most recent additions to the MCU has enabled Marvel to pursue more experimental styles, evidenced by series such as WandaVision, Moon Knight and Ms. Marvel. As such, it is not uncommon for these shows to be met with a broad range of opinions. In particular, Ms. Marvel has seemed to trigger a very polarized response, with many claiming that the show has been review-bombed for various reasons. Using the tools of multiparadigm data science, I set out to examine this claim and see what (if any) robust statistical evidence can be found to support it. The framework I end up creating is very general, and facilitates the collection and analysis of rating data on games, TV shows and movies more broadly.

Deploying Standalone Wolfram Language Applications

Chris Cooley Watch Video

What's New in ExternalEvaluate

Riccardo Di Virgilio Download Presentation Watch Video

The ExternalEvaluate framework lets you evaluate code from a number of external languages and environments such as Python, Ruby and Java directly in your Wolfram session. This talk will focus on the latest additions, with particular attention to the recently added SQL, shell and Julia components.

Introduction to Graph Neural Networks and What We Can Implement in the Wolfram Language

Mike YehWatch Video

I will give a quick review of the ideas of graph neural networks (GNN), then overview the potential types of GNN and show more details of algorithms that can be implemented in the Wolfram Language. I will then show how to implement these types of GNN in the Wolfram Language.

Locality & Bell Theorems

Michael L Ulrey Download Presentation Watch Video

It is often said that Bell's theorem demonstrates the impossibility of "local realism" as a basis for explaining quantum mechanics. In this talk, we attempt to separate the hype from the actual mathematics. A simple operational (i.e. statistical) modeling framework is introduced that includes the quantum mechanics explanation of the famous Einstein–Podolsky–Rosen–Bohm (EPRB) experiment as well as certain so-called classical alternatives. Within this framework, it is easy to show that each of the classical alternatives fails (in possibly different ways) to reproduce at least some of the statistical behaviors predicted by quantum mechanics for EPRB. But at this level, it is purely mathematical assumptions that distinguish one model from the other. How do wild statements about the nonlocality of the universe enter the picture? The goal is to identify the central idea that clarifies what locality, determinism and realism are all about in this context.

Spatial Values—Spatial Prediction

Eduardo Serna, Gosia Konwerska Download Presentation Watch Video

Spatial datasets consisting of a set of measured values at specific locations are becoming increasingly important. Examples include temperature, elevation, concentration of minerals, etc. We will look at existing Wolfram Language functionality to perform estimation of missing values in a region of interest, modeling the underlying quantity. Traditionally, this procedure has been known as kriging.

Spatial Events—Spatial Statistics

Eduardo Serna, Gosia Konwerska Download Presentation Watch Video

Spatial point patterns are collections of randomly positioned events in space. Examples include trees in a forest, positions of stars, earthquakes, crime locations, animal sightings, etc. Spatial point data analysis, as a statistical exploration of point patterns, aims to answer questions about spatial randomness, point density and interpoint interactions. In this talk, we will explore the existing Wolfram Language functionality designed for the analysis and modeling of spatial events.

Text Summary Model with BERT

Mike Yeh Watch Video

In the beginning, the text summary task and the basic properties of the BERT model will be reviewed. Then I will introduce the idea and the algorithm of the text summary paper that we are going to implement. Finally, I will show the steps of building the text summary model by importing model weights.

Exploring Chemistry with Wolfram Language Paclets

Jason Biggs Download Presentation Watch Video

This talk presents a set of paclets available from the new Wolfram Paclet Repository that extend the chemistry functionality in the Wolfram Language. Exhaustively generate isomers from a chemical formula using the MAYGEN or surge libraries, perform semi-empirical quantum chemistry using MOPAC, generate new and novel compounds using SELFIES strings and connect to the Open Reaction Database.

What's New in Chemical Education

Camilo Castrillon-Toro, Jason Sonnenberg Download Presentation

Recent and upcoming features of Wolfram|Alpha, Wolfram|Alpha Notebook Edition and the Wolfram Language related to chemical education are discussed. These features include expanded step-by-step solutions support, improved free-form input, new built-in functions for chemical conversion and new Wolfram Function Repository submissions. Example activities and educator resources are highlighted.

Simplified Machine Learning Workflows Overview

Anton Antonov Download Presentation Watch Video

In this presentation, we overview the main Machine Learning (ML) workflows and their rapid specification through monadic packages and a Natural Language Processing (NLP) template engine. We demonstrate that the ML workflows we consider are universal—they have multi-language implementations in Python, R, the Wolfram Language, and others. The overview covers Data Wrangling, Classification, Regression, Latent Semantic Analysis, and Recommendations. Extensive explanations and examples are given in the book Simplified Machine Learning Workflows (which should be finished before the conference).

Building Your Own UIs: Tips from the Trenches

John Fultz, Lou D’Andria, Kevin Daily

What's New for Physics Education: Formulas and Quantities

Jason Martinez, Daniele Ceravolo Download Presentation Watch Video

Wolfram|Alpha provides step-by-step explanations for solving formulas and operations involving quantities and, through natural language processing, representations for those queries in the Wolfram Language. Explore recent and upcoming improvements to step-by-step functionality such as unit choice for results, support for multiple equations, expanded step descriptions and equation derivations that explain the physical principles involved. Also, learn about recent improvements in Wolfram Language interpretations such as supplying default values for system choices and Wolfram Function Repository interpretation coverage.

Mastering Fine Arts with Mathematica: Using Code as Artistic Medium

Jack Madden Download Presentation Watch Video

In the Spring of 2020, I defended my Ph.D. in astrophysics knowing that, in the Fall, I would be headed to art school. Having no formal training in traditional mediums like painting, drawing or sculpture, I relied heavily on the creative versatility of Mathematica and my years of using it for scientific research to get a Master’s in Fine Arts. My work focused on creating programs that explored the human connection to science and the challenges of internalizing astrophysical discovery. Mathematica showed itself as a powerful tool for creating art and rapidly iterating toward a more subjective end result than I was used to. My success in developing award-winning code art has encouraged me to teach Mathematica to artists and designers at the Rhode Island School of Design this Fall. In this talk, I will go over my journey into art through the use of Mathematica and how the current art world has embraced code as a creative medium. I will discuss new ways I've learned how to use Mathematica for generating art and highlight some of my students' work as "artist first" Mathematica users.

Office Hour: System Modeling

Niko Era-Esko, Sergio Vargas, Jan Brugard, Suba Thomas, Vedat Senol

Do you have any questions related to System Modeler or modeling in general? Meet us during our office hour and we will be happy to answer your questions.

Book Publishing with the Wolfram Language

Paige Bremner, Treyton Jansen DownloadPresentation Watch Video

Learn about book publishing opportunities with Wolfram Media and beyond. We will provide an overview of Wolfram Media, the publishing unit of the Wolfram group. We'll explain our various programs for supporting authors working to publish both with Wolfram Media and with other publishers. A brief demonstration of the Wolfram Book Tools palette for authoring will be given. We will leave time for Q&A, which will be an excellent time to report any publishing-related bugs you may be experiencing. This workshop is designed for both authors already working on a publishing project with Wolfram Media or another publisher and anyone who is interested in writing a book that serves to promote Wolfram technology.

Calling External Libraries with the Compiler

Christopher Wolfram Download Presentation Watch Video

Learn about new features of the Wolfram Language compiler that make it possible to efficiently interact with external libraries. LibraryFunctionDeclaration makes it possible to call library functions in compiled Wolfram Language code, while functions like ToRawPointer, CreateTypeInstance and InertExpression make it possible to efficiently convert between expression-based evaluated code and library-specific data structures.

Office Hour: Probability & Statistics

Eduardo Serna, Gosia Konwerska

Drop in and chat with our experts. Ask questions, get help, or just say hello!

OpenAI GPT-3 and Wolfram

Maureen Baehr Watch Video

StartingUpGood is a collaboration whose purpose is to support fresh entrepreneurial approaches to do more good in the world. Our core areas of social impact interest are:

  • Sustainable Development Goals—data and counting.
  • Gathering and curating information on startups, impact investing and social change.
  • Implementing projects to improve social enterprise representation and funding equity.

We've been in existence for 7 years, using Wolfram as our central organizing software and attending and presenting at conferences. Because much of our work is based on finding, curating, extracting, analyzing and formatting information, Wolfram is the valuable hub for organizing our work. Since the start, it has been useful for structured data, and more recently we've been using enhanced tools for PDF, video, WebSearch, etc. We've also incorporated it with other tools—Excel, Word, etc.—to find the best hybrid solutions for various projects. We've also explored using Wolfram machine learning for certain tasks. When OpenAI announced the beta for their GPT-3 models last year, we applied and were accepted. At first we used the Playground for small experiments, but once we found Daniel Bigham's great code for incorporating the OpenAI GPT-3 API into Wolfram code, we began to use it for more formal work. The presentation will show several examples of how we've incorporated hybrid Wolfram/OpenAI GPT-3 into our workflows and will consider opportunities and constraints.

Overview of CodeTools: What's New Analyzing and Developing Wolfram Language Code

Brenton Bostick Download Presentation Watch Video

This overview will include a tutorial on setting up VS Code to develop Wolfram Language code, show what's new in the front end for analyzing and formatting code, and show how to manipulate Wolfram Language code symbolically with CodeParser.

Meet-Up: Deploying Standalone Wolfram Language Applications

Chris Cooley Download Presentation

Have you ever wanted to share your Wolfram Language code with friends and colleagues who do not have the Wolfram Language installed? Have you ever wanted to sell your code to the public? At Wolfram Research, the ongoing standalone applications project is working on making it easy for Wolfram Language developers to deploy their paclets, scripts and other code into application binaries that can be executed anywhere, by anyone even when the Wolfram Language is not installed. In this talk, we will describe the status and goals of the standalone applications project. Primarily, we will discuss our proposed technical process for generating robust, practical applications from Wolfram Language code.

Social Engagement in Online Learning

Clinton Bradford Download Presentation Watch Video

Student-to-student interaction, competition, and communication have been used to great effect in in-person learning. However, online learning platforms have lagged behind in their ability to use these effective educational tools. At SYLVA, we are creating tools for instructors to design, implement, analyze, and run synchronous and asynchronous polling, interaction, and competition experiences for education and research. In this talk, we will go over some of the possibilities this technology can enable and some of the technical solutions needed in order to allow instructors to run no-code and low-code experiences at a modern online-education scale.

Instrumenting the Wolfram Language: Profiling, Coverage and Debugging

Brenton Bostick Download Presentation Watch Video

Instrumentation` is a new CodeTools package that automates instrumentation of Wolfram Language code for common software engineering tasks such as profiling and code coverage reporting. This talk will show how to use Instrumentation` to profile your code and will also talk about future directions for Instrumentation` that include new debugging technology.

Model Transformation for Smart Grid Applications

Glen Halley Download Presentation Watch Video

With the current interest in alternative energy sources, electric utilities are facing challenges to understand their behavior when integrated into the network. One approach used to gain understanding during design and operation is the use of simulation. Most simulation tools used by electrical utilities are "black boxes" that do not allow users to easily extend their analysis with new components or to take advantage of modern computational tools, locking the models in a single environment. To bring utility models into Modelica and the Wolfram ecosystem, there needs to be an approach that allows import of existing models into Wolfram’s System Modeler. In this presentation, we will present OpenIPSL, a library of electrical grid modeling components built using Modelica that works with System Modeler to conduct time-domain simulations. While this library provides the building blocks for modeling, transforming models requires a transformation tool. We will introduce a prototype transformation tool and demonstrate how it can be used to bring models into System Modeler. The demonstration includes the translation of both PSS/E and CIM model files into System Modeler and a discussion of the verification and validation of the translations. Translation prevents the need to recreate the model and can more rapidly solve problems. The model is easier to understand for PSS/E users because it is a one-for-one translation. This translation tool could be the starting point for many model sources.

Office Hour: Geo & Astro

Jose Martin-Garcia

Drop in and chat with our experts. Ask questions, get help, or just say hello!

City-Scale Hybrid Simulation COVID-19 Spread

Charles Macal Download Presentation Watch Video

As the COVID-19 pandemic continues to unfold, the epidemiological modeling community continues to support decision makers with forecasts of how the disease could progress under a variety of assumptions: the emergence of COVID variants; the public's response to resurging COVID-related cases, hospitalizations, and deaths; new vaccination regimes; and potential interventions by decision makers. In this talk, we describe two complementary Mathematica epidemiological simulation modeling approaches, one an agent-based model (ABM) and the other a compartmental model based on System Dynamics (SD), which have been developed to forecast COVID-19 spread at the city scale, specifically for Chicago. The COVID ABM simulates the movements, contacts, and disease transmission outcomes of every person in a city and installation of facility as they go about their daily activity schedules. Complementing the agent-based model, the System Dynamics COVID model segments the population, compartmentalized by current disease state. Used together, the hybrid system combining the top-down SD model with the bottom-up ABM provides a broader understanding of disease spread than is available with either approach alone. The multiparadigm modeling capability of Mathematica is found to easily and elegantly express essential formalisms of agent-based and System Dynamics modeling. The talk concludes on how agent-based modeling and System Dynamics can work together to best inform public health decision-making.

Workshop: ExternalEvaluate

Riccardo Di Virgilio Download Presentation Watch Video

Come to a hands-on session with the developers of the ExternalEvaluate framework. Bring your questions, suggestions, ideas and problems while working with external languages. Don't have a specific question? Then try your hand at one of our challenges in Java, Python, JavaScript, Julia or SQL.

Correlative Analysis for Instrumentation Used in Atmospheric Chemistry

Colleen Marciel Rosales Download Presentation Watch Video

Correlative analysis of environmental data, such as those acquired from long-term air quality monitoring networks across the US, requires tools that allow one to simultaneously look at trends in multi-parameter chemical analyses, one example of which is the multi-elemental characterization of particulate matter (PM2.5) samples in air. In this talk, the correlative analysis of elemental concentration in PM2.5 acquired using two different chemical instruments are detailed. The first part will be focused on data organization using the Dataset functionality. The second part will explore the visualization and correlation calculations (including linear regression, reduced major axis regression, and Theil–Sen regression) on PM2.5 data and how these tools are useful in evaluating the performance of the instruments (1) as an internal check, and (2) relative to the other instrument.

Thursday, October 20

Water Resources Management with the Wolfram Language

Ricardo Martinez-Lagunes Download Presentation Watch Video

Integrated Water Resources Management requires the analysis of information from multiple sources. Data must be visualized for different audiences.

Notebook Interface for Distributing Mathematica Programs to End Users

Mike Grigola DownloadPresentation Watch Video

CB&I is the world's leading designer and builder of industrial storage facilities, tanks and terminals. We use Mathematica to perform the engineering calculations to design our products. The functions that run these calculations may have hundreds of inputs that require proper units, ranges, and data types. We need to help users quickly enter and validate these inputs and fix problems such as accidentally modified or deleted code while keeping the flexibility that Wolfram Notebooks allow. Our solution is a package called NotebookInterface. The package combines the benefits of standard Input cells that can be modified by the user and a dynamic user interface that validates inputs, highlights changes, and repairs invalid or out-of-date cells. We will show how we use the package in a large corporate environment and how you can use it to create user interfaces for your own software.

The Wolfram Compiler

Tom Wickham-Jones Watch Video

The Wolfram Compiler converts Wolfram Language code into native machine instructions that run directly on the machine hardware. It offers a powerful way to speed up programs and provides many opportunities for innovative programming features such as building connections to external libraries and accessing new technologies. It is also used for an increasing amount of new development. This talk will describe the Wolfram Compiler and its latest advances and directions.

What's New in Wolfram Video

Shadi Ashnai Download Presentation Watch Video

In this talk I will show the state of the Video object in the Wolfram Language as well as all the computational power developed around it in order to create, process or analyze videos. https://wolfr.am/video

Neural Networks for Audio Effects

Carlo Giacometti, Rebecca Frederick Download Presentation Watch Video

Extreme Citizen Science—Intention Building, Rewarding Risk-Taking

Colin Chapman Download Presentation Watch Video

The Victorian Curriculum and Assessment Authority Systems Engineering Study involves the design, iteration and production of integrated systems. Integral to the study design is the identification and quantification of systems goals, the generation of system designs, trial and error, justification, modelling and the implementation of a response. The study considers the interactions of systems with people, society and ecosystems and seeks to optimise system efficiency and performance through agile processes. Learners design, conduct, analyse, critique and display meaningful research. The study design rewards risk taking with rubric-assessed processes of intention development, making and evaluation of response, and iteration processes. The study encourages multimedia narratives that recognise those moments when intentions met reality. Wolfram System Modeler and Mathematica through the Computational Essay approach allow rich, thick descriptions of achievement as learners engage in responses to curated stimulus in an Extreme Citizen Science frame. This workshop works through the development of a Polarimeter from scratch using an Arduino microcontroller, the ModelPlug library in Wolfram System Modeler and the computational functionality of Wolfram Mathematica. The Computational Essay gives a narrative of data wrangling, uncertainty and calibration. The Polarimeter was contrived to determine the sugar content of soft drinks, in support of an investigation into athlete nutrition.

Software Engineering and Computer Science with the Wolfram Compiler

Tom Wickham-Jones Watch Video

The Wolfram Compiler is one of the largest applications built from the Wolfram Language. This talk will review the software engineering practices used to build the Wolfram Compiler, ensure it runs correctly and improve its efficiency. It will also discuss its internal organization, showing how this is an innovative way to understand compiler-related computer science. It will discuss how the compiler is a powerful way to learn about compilers and also to develop new compiler techniques.

Fun Applications in Image Processing

Markus van Almsick Download Presentation Watch Video

Office Hour: Wolfram Language Chemistry

Jason Sonnenberg, Jason Biggs, Robert Nachbar

Come chat with the chemists who develop chemistry in the Wolfram Language.

Programming with Data Structures

Tom Wickham-Jones Watch Video

The Wolfram Data Structure library offers performant implementation of many traditional data structures. It is implemented with the Wolfram Compiler. This talk will review the current data structures and show how they can be used to improve your code. It will also show how data structures can be used inside compiled code to gain particularly optimal performance.

What's New in Wolfram|Alpha for Mobile

Aditya Pabbaraju, Nick Zitzmann Download Presentation Watch Video

Wolfram|Alpha on mobile devices has changed this year. In this presentation, we will talk about everything that is new with Wolfram|Alpha for iOS and everything that is to come soon to Wolfram|Alpha for Android. We'll be going over the new look and the new features, including math OCR and the personal and professional assistants.

Should I Eat That? Food Science and Safety with the Wolfram Language

Isabel Skidmore, Gay Wilson, Tommy Peters Download Presentation Watch Video

This talk will cover the structure and innovative uses of food data within the Wolfram Language. We will demonstrate how to retrieve data for popular fast foods and how to analyze and manipulate a large dataset of alcoholic beverages. We will introduce new food safety data available in the Wolfram Language, including maximum refrigerator and freezer storage times for more than 200 foods, and demonstrate the synergy of Wolfram|Alpha food data and biology data to analyze foodborne-illness pathogens and their food sources. Next, we will invite the audience to play "Should I Eat That?," in which audience members have fun testing their food safety knowledge against Wolfram|Alpha data. We'll wrap up with a sneak peek at exciting future developments for food in the Wolfram Language.

The Art of Problem Solving in Wolfram Language in '22

Shenghui Yang Download Presentation Watch Video

In '20 I focused on Olympiad geometry and in '21 I covered four problems about circle inversion, asymptotics, number theory and combinatorics just like the IMO exam syllabus. Both talks were welcomed by virtual event attendees and YouTube viewers. This year I am going to use the same instruction style as in '21 and demonstrate problem-solving skills in these four domains. In the past years there have been a large amount of Wolfram Function Repository submissions, which opens several new paths for solving challenging problems with even fewer lines of code and in clearer ways of thinking. I have selected couple of problems and examples from the following published resources: L.A. Graham's Ingenious Mathematical Problems and Methods, R. Honsberger's From Erdös to Kiev: Problems of Olympia Caliber (or related CRUX volumes), M. H. Weissman's An Illustrated Theory of Numbers and lastly Stanley's Enumerative Combinatorics. The Wolfram Language's built-in functions and Wolfram Function Repository resources have a unique charm to make the exploration of the solutions very expressive. It is critical to see the connection between the "Aha!" moment and solid examples generated by the Wolfram Language, even with silly brute-force methods at first. It is my wish all my audience will enjoy the talk and start their own journeys of solving challenging problems with the Wolfram Language.

The State of Graphics in the Cloud

Matthew Adams Download Presentation Watch Video

This talk gives an update on our efforts to bring the full power of Wolfram Language graphics to the Wolfram Cloud.

Automated Planar Geometry

Dan McDonald, Peter Barendse, Jack Heimrath, Jeremy Stratton-Smith Download Presentation Watch Video

We present updates to the automated geometric functionality of the Wolfram Language introduced in Version 12, including new functionality for automated geometric reasoning and for creating GeometricScene objects.

New Biology Content in the Wolfram Language

Keiko Hirayama, Lina Marcela Ruiz-Galvis, Carlos Munoz Amezcua Download Presentation Watch Video

Wolfram technology provides access to a diverse range of computable knowledge about biology. In this talk, topics such as taxonomic classifications, dinosaurs and genomics are explored as examples of the breadth of information available in the Wolfram Language. The latest taxonomic data collections deliver biological classification of groups of organisms and the ability to analyze them by physiological, geographic and ecological characteristics. Furthermore, new standardized dinosaur datasets are introduced to compute information such as discovery locations, geological periods and formations, behaviors, and diets, providing answers to the most popular dinosaur queries. Finally, updates to genomics data include access to gene homology information, as well as tying gene properties like chromosome position to the latest reference genome assemblies, while maintaining data for previously supported human assemblies. Genome assembly statistics, such as release date and percentage of gaps, are also included.

The State of the Wolfram Cloud

John Pacey Download Presentation Watch Video

After a general introduction to the Wolfram Cloud, we'll talk about what's new in the world of cloud notebooks and beyond. We'll also give a brief update at the ongoing "containerization" of the cloud architecture, which will be particularly interesting for Wolfram Enterprise Private Cloud (EPC).

Office Hour: Visualization

Brett Champion, MinHsuan Peng, Nirmal Malapaka, Tim McDevitt, Yuzhu Lu

Drop in and chat with our Visualization team. Ask questions, get help, or just say hello!

What's New in Step-by-Step Mathematics

Jordan Hasler, AnneMarie Torresen, Judy Zhang, Jack Heimrath, Jeremy Stratton-Smith, Carlos Andres Ballesteros-Alvarez Download Presentation Watch Video

An introduction and review of the latest step-by-step programs for Wolfram|Alpha including a discussion of developing step-by-step functions from both algorithmic and rule-based approaches will be shown. Topics to be discussed will include new steps for elementary number theory, grade-school algorithms for addition and multiplication, trigonometric simplifications and expansions, statistics, vector analysis, difference equations, and recurrence relations.

Introducing Wolfram Web Engine

Carlo Barbieri Download Presentation Watch Video

Wolfram Web Engine is a new solution for serving your Wolfram Language–based web applications. It is lean and mean and you can configure it any way you want. It supports modern Wolfram Language web deployments like APIFunction and FormFunction, and is also backwards compatible with JSP-based technology. You can run it directly on your own server, or deploy it in a Docker container. When you want to give the power of the Wolfram Language to the world, this is your gateway.

Topical Meet-Up: Notebook Front End

John Fultz, Ian Hojnicki, Lou D'Andria

An informal gathering for all who are interested in sharing insights and experiences, asking and answering questions and connecting with others for an in-depth discussion.

Workshop: Working with Relational Databases in the Wolfram Language

Carlo Barbieri Download Presentation Watch Video

The objective of this workshop is getting you started with the main concepts of working with Relational Databases through the Entity framework in the Wolfram Language.

Function Extraction: Mapping Programs into Mathematica Equations

Ali Mili Download Presentation Watch Video

Much of the trouble in software engineering stems from our inability to determine precisely the function of a program. Programmers routinely write programs whose precise function they do not know, and can only speculate about, and most modern software development processes involve composing existing software components on the basis of documentation that is typically vague, incomplete and outdated. In this talk we discuss the design and implementation of a tool that derives the function of a C-like program by mapping it onto a mathematical equation formulated in the syntax of Mathematica (© Wolfram Research). Such an equation involves the program’s inputs and outputs; by solving this equation in the output as a function of the input, we obtain the function of the program. Also, whether the function can or cannot be solved, it can still be used to prove or disprove the correctness of the program with respect to specifications. The unique novelty of our approach is that we can capture the semantics of iterative constructs in full, provided we codify the requisite programming knowledge and domain knowledge.

Multispectral Astronomical Imaging

Tom Sherlock Download Presentation Watch Video

Using Wolfram Language, many aspects of astronomical observations can be automated. Using Wolfram Language and .NET/Link to interface to the industry standard ASCOM interface, a monochrome CCD camera and a motorized filter wheel were controlled to automate the tedious process of capturing an image sequence through four color filters plus calibration frames. All the alignment, stacking and calibration plus the initial image processing were done in the same notebook using Python routines called from Wolfram Language code.

New in System Modeler

Jan Brugard Download Presentation Watch Video

In this presentation, the key features for System Modeler 13.1, as well as the upcoming release, will be presented. This includes new free model libraries, task-oriented documentation and integrated machine learning workflows.

Office Hour: Notebook Front End

John Fultz, Ian Hojnicki, Lou D'Andria

Drop in and chat with our experts. Ask questions, get help, or just say hello!

Model Predictive Control

Suba Thomas, Sergio Vargas Download Presentation Watch Video

This talk will showcase the recently added functionality to design model predictive controllers. The formulation of the problem as a multiparametric optimization problem, along with workflows and related functionality, will be presented.

A Computational Study of the Internal Structure of Historical Papers: Leonardo da Vinci's Notebooks

William Sethares Download Presentation Watch Video

Leonardo da Vinci's manuscripts present a puzzle for scholars who aim to recreate the pages' original formats, sequence, and chronology. While the most practical way to capture just two inner structural features—watermarks and chain lines—is via transmitted light, these traces of manufacture can be difficult to decipher when writing and other surface marks found on the recto and verso of the sheet obscure them. Using diffuse specular (normal) light and transmitted light images of Leonardo's Codex Leicester (Bill Gates Collection) and the Codex Arundel (British Library) as case studies, we will describe a computational approach, whereby surface marks and writing are minimized in order to facilitate the accurate matching of watermarks and chain/wire lines. Using enhanced images of the inner structural features of da Vinci's papers, it is then possible to suggest moldmates (papers made from the same papermaking mold) in both Codices. The existence of moldmates in one or both manuscripts provides meaningful evidence of origin, sequence, and date.

Cohesive Games and Lessons Learned from the Field Theory of Games

Gerald Thomas Download Presentation Watch Video

Based on teaching the Field Theory of Games to upper-class students, there were interesting lessons learned. For example, it is possible in a short time to educate them to use the Wolfram Language sufficiently to apply the field theory of games to problems of interest to them. It is possible to teach them the elements of traditional game theory and the key elements of the field theory of games in that same time. However, there were ideas that appeared more difficult to communicate: changing the mindset that decision processes have a fixed stable behavior as opposed to a dynamic behavior. This talk explores these issues and focuses on an idea that is important and was not communicated in the course: games can be cohesive and thus illustrate concepts of control and effectiveness.

Neural Networks in System Models

Jan Brugard, Sergio Vargas Download Presentation Watch Video

In this presentation, you will see how data provides new avenues and connections to physical modeling with the new support for neural networks in system modeling. Applications and examples will be explored.

Simulating the Hydrodynamic Nature of Porosity

August Young (Frechette) Download Presentation Watch Video

The effective porosity of a medium defines the volume of pore space conducive to through-flow (otherwise known as the "mobile zone"). In the field of groundwater hydrology, the concept of effective porosity is well known, but there exists no such work in the literature that details the dependence of effective porosity on the Reynolds number. Knowing the details of the relationship between the Reynolds number and effective porosity affords researchers the ability to manipulate the effective porosity of a medium. Additional benefit comes from the ability to more accurately estimate rates of mass transfer between cavities and mobile zones, which directly impacts the calculated timescales, costs and therefore the anticipated feasibility of groundwater remediation projects. To determine the effective porosity of a porous medium, we study the medium at pore scale and assume an idealized pore geometry that separates the pore space into two regions: the "immobile" and mobile zones. The hydrodynamic boundary between these zones is known as the "separatrix." To observe movement of the separatrix in the idealized pore space, we use Mathematica's numerical differential equation solver, NDSolve, to solve the Navier–Stokes equations and associated boundary conditions. The suite of tools built into the Wolfram Language allows for straightforward postprocessing of the resolved flow field, which we demonstrate through use of the DynamicModule and StreamPlot functions to describe the position of the separatrix, and therefore the effective porosity of the medium, under various flow conditions.

Using Paclets from the Wolfram Language Paclet Repository

Todd Gayley Download Presentation Watch Video

This talk will demonstrate how to install and use paclets from the new Wolfram Language Paclet Repository. Paclets are the ideal way to package and distribute significant pieces of Wolfram Language functionality. The Wolfram Language Paclet Repository is a growing collection of useful add-ons contributed by users and developers like you. This presentation will focus on the basics of paclets and how to use ones from the Wolfram Language Paclet Repository; later talks will dive into how to develop your own paclets for submission. Come learn how to use and contribute to this exciting new repository!

Get Started Developing Paclets for the Paclet Repository

Bob Sandheinrich Download Presentation Watch Video

The Wolfram Language Paclet Repository is here! This talk shows how to use the paclet definition notebook to help develop a paclet and prepare it for publication or distribution.

Compositional Structure of Classical Integral Transforms

Oleg Marichev Download Presentation Watch Video

The recently implemented fractional order integro-differentiation operator, FractionalD, is a particular case of more general integral transforms. The majority of classical integral transforms are representable as compositions of only two transforms: the modified direct and inverse Laplace transforms with power multipliers. Using MellinTransform, one can construct two main classes of integral transforms: convolution type (FourierSinTransform, FourierExpTransform, HankelTransform, Stieltjes transform, G-transform, etc.) and transforms with respect to indices (Kontorovich–Lebedev transform, Mehler–Fock transform, Wimp transform, etc.). For these transforms, we give definitions, inversions, compositional structure, representations through more general transforms (like the G- and Wimp transforms) and more. These transforms can be directly applied to rational and other functions representable through MeijerG functions.

Continuous Integration & Deployment of Paclets

Richard Hennigan Download Presentation Watch Video

Documentation Tools/Authoring Documentation for Paclets

Brian Van Vertloo Download Presentation Watch Video

Office Hour: Wolfram|Alpha Notebook Edition

AnneMarie Torresen, Brad Janes, Jason Sonnenberg, Jeremy Stratton-Smith, Jordan Hasler

Come chat with the educators who develop Wolfram|Alpha Notebook Edition.

Office Hour: Sound & Vision

Carlo Giacometti, Shadi Ashnai, Piotr Wendykier, Markus van Almsick

Drop in and chat with our experts. Ask questions, get help, or just say hello!

What's New in Geometry

Charles Pooh Download Presentation Watch Video

Progress in Graphs & Networks

Jaebum Jung Download Presentation Watch Video

Extending the Front End with LibraryLink

Christopher Cole Download Presentation Watch Video

Region Reconstructions: Uses and Techniques

Alec Shedelbower Download Presentation Watch Video

Cell Tray Widgets

Jason Abernathy Download Presentation Watch Video

The cell tray widget interface allows developers to produce high-quality cell-level user interface tools that are managed by the front end. Widgets are organized into "trays" placed near the cell bracket.

Office Hour: Wolfram Cloud

John Pacey

Drop in and chat with John about our cloud development. Ask questions, get help, or just say hello!

Prostate Antigen Kinetics

Alan White Download Presentation Watch Video

Differential equations are solved to the task of graphing the change in concentration of the prostate-specific antigen in prostate cancer patients. Note: Most equations are original to the writings of Ernest Rutherford.

Implementation Story: The Notebook Toolbar

Noah Hardwicke Download Presentation Watch Video

Implemented in top-level Wolfram Language code, the new notebook toolbar used existing features of the desktop front end to produce an interface with unique structure and function. In this talk I will review the front end technologies that were used and detail the construction of the toolbar, especially how attached cells provided a viable way of efficiently eliding controls into an overflow menu. This talk will be of particular interest to those wanting to build attached cell-based interfaces in the Wolfram Language.

Biological Data Insight & Modeling

Mark Kotanchek Download Presentation Watch Video

Multiomics (genomics, proteomics, lipisomics, etc.) data is typically ill-conditioned with many (coupled) variables and relatively few data records. As a result, such data sets are very difficult to analyze with conventional statistical and machine learning techniques. ParetoGP assumes we can develop simple algebraic models using just a few of the inputs via an evolutionary search rewarding model simplicity and accuracy. From the thousands of candidate models, we can garner insights on key variables, variable associations and combinations as well as generate concise explainable and human-interpretable models. In this presentation, we demonstrate the insight and modeling process using DataModeler applied to cancer therapeutics and bone regeneration data sets.

Cambodia Wolfram Student Stars! Mathematica, Arduino, Raspberry Pi

Linus Anaka Download Presentation Watch Video

The Cambodia Ministry of Education, Youth and Sport, with support from Wolfram Research, invited students in grades 4–12 to participate in Cambodia Wolfram Student Stars! In addition to Wolfram computational thinking activities, we supported student STEM projects with System Modeler and Mathematica using Arduino and Raspberry Pi. As CWSS STEM project director, I will describe the implementation of the program in Cambodia to support STEM projects, and share student STEM projects.

Friday, October 21

13.2 Graphics

Yuzhu Lu Download Presentation Watch Video

This talk will present some new features and updates of Mathematica's 2D and 3D Graphics language, including DropShadowing, Blur, improved support of AbsoluteOptions, 3D Camera dolly and look-around interaction, and other improvements. This talk will also show users how to customize 2D/3D rendering with SurfaceAppearance.

Cellular Automata Rule-Generating Polynomials

Rodrigo Obando Download Presentation Watch Video

Cellular Automata rules are represented by integers where we encode the output of the function without knowing the details on how it might be implemented. The CellularAutomaton function in Mathematica only requires these integers, along with the values of r and k, to evolve rules for a given input. The BooleanFunction function in Mathematica provides a polynomial of the function given its representing integer which we can use to get the functions when k = 2. Unfortunately, there is no Mathematica function that does the same for other values of k. We present a uniform way to generate these functions using certain kinds of polynomials. Every function in a rule space can be generated by changing the coefficients of its corresponding polynomial. We find that when the value of k is prime, the polynomial is relatively simple and straightforward. When the value of k is composite, we need a composition of polynomials, each based on a prime factor of k. A possible use of these polynomials is the creation of Probabilistic Cellular Automata by defining one or more of the coefficients as random variables. For every step, for every cell, the random coefficients will take on different values and hence generate different rules. This may allow for a better analysis of these stochastic processes.

Office Hour: Geometry

Jaebum Jung, Charles Pooh, Yan Zhuang, Alec Shedelbower, Keren Garcia

Drop in and chat with our experts from the Geometry team. Ask questions, get help, or just say hello!

Virtual Origami

Tetsuo Ida Download Presentation Watch Video

Origami attracts the minds of people all over the world. Some are interested in its geometric aspects, and others in artistic or recreational elements, so-called traditional origami. Although both origami categories rely on a single notion of paper folding, their methodologies differ due to the pursuits' objectives. We focus on traditional origami, but approach it with mathematical formalism rather than artistic values. In particular, we realize origami in a virtual space, using Mathematica's leading-edge symbolic, numeric, and graphics computation capabilities. The origami creators' interest lies more in artistic creativity than general folding rules. In contrast, origami geometers' underlying motivation is finding a few basic fold rules. Although we have such a set of rules in 2D origami geometry, we are more liberal in defining basic fold rules in traditional origami. The fold rules may be many, even open (infinite). Nevertheless, we can give some basic fold rules essential in traditional origami. We first address squash and outside-reverse folds among classical folds; these classical folds intricately combine several simple folds. One squash fold, for example, requires three simultaneous simple folds, i.e. mountain or valley folds, subjected to nontrivial constraints. We understand the 2D origami geometry to the extent that we do the 2D Euclidean geometry in terms of the constructible points in the two geometries. Namely, the set of the points constructible by Huzita–Justin's fold rules is the strict superset of intersecting points of the circles and the lines made by a straightedge and a compass, the construction tools of Euclidean geometry. We extend Huzita–Justin's rules by introducing a cut along a crease. The operations of cutting and later gluing simplify a squash fold significantly. We can apply the cut operation to other classical folds and streamline their modeling and subsequent realization in a virtual origami space. We then present the construction of a flying crane. It not only serves as a mathematical and computational description of this famous origami but also shows the effectiveness and expandability of our modeling for origami virtualization.

Automated Geometry Theorem Generation

Philip Todd Download Presentation Watch Video

We present a method for automatically generating new geometry theorems, suitable for solution using "angle chasing". Specifically, we consider geometry theorems whose premises and conclusion comprise a set of bisector conditions. Each premise and the conclusion can be represented as the rows of a "bisector matrix": one whose rows contain 1,1,–2 as their only nonzero entries. The existence of a theorem corresponds to rank deficiency in this matrix. We initially consider rank-deficient matrices which have an underlying graph whose vertices correspond to matrix rows and whose edges correspond to matrix columns. We give each edge of the graph (considered in a specific direction) a head-weight equal to the matrix entry in the row corresponding to the vertex adjacent to its head and column corresponding to the edge. Similarly, we give the edge a tail-weight. With this definition, a bisector matrix with an underlying graph is rank deficient if and only if the graph is bicubic and for any cycle the product of head-weights equals the product of tail-weights. We present a Mathematica program which uses this characterization to identify all possible such matrices for 14 or fewer rows. We present methods of transforming the rank-deficient matrices into forms which incorporate line angles in addition to bisectors, and which have columns of more than two non-zero entries, while retaining the rank deficiency. We present code which takes such transformed matrices as input and outputs geometry theorems in the form of GeometricScene.

Everything Arrays

Rob Knapp Download Presentation Watch Video

Working with arrays is fundamental to efficient use of Wolfram Language and appears in almost any computation. The basic representation for an array is a nested list, but there are other representations such as PackedArray, SparseArray and structured arrays that are more effective for some applications. This presentation will show how to use the various representations of arrays combined with compilation to make your computations more efficient.

The Structure of Paclets

Connor Gray Watch Video

Wolfram Optimization

Paritosh Mokhasi Download Presentation Watch Video

This talk will give an overview of the various optimization functions that can be used to solve a wide variety of convex, nonconvex and multidomain problems. The Wolfram optimization functionality will be demonstrated using a diverse set of examples. Visiting this talk will enable you to solve optimization-type problems efficiently.

Classical and Quantum Subjectivity

Stuart Nettleton, David Boyle Download Presentation Watch Video

Uncertainty is a major component of subjective logic beliefs. We discuss the cloud of uncertainty across Markov networks, insights from computational irreducibility, and negative quantum quasiprobabilities and beliefs.

Meet-Up: Virtual Labs in Education

Ankit Naik, Jan Brugard

In this meet-up, we will discuss the virtual labs, including ideas for new labs.

Global Optimization

Nina Dokeva Download Presentation Watch Video

This talk will discuss the global optimization functionality in Wolfram Language. It builds upon the convex and convertible to convex optimization functionality developed previously and extends to functionality for solving nonlinear, nonconvex problems with real-valued and mixed-integer variables. Available methods and library solvers will be discussed and illustrated by examples.

Natural Language Processing Template Engine

Anton Antonov Download Presentation Watch Video

In this presentation we discuss the completion of computational templates with parameters that are extracted from text specifications using a question answering system (QAS) [Wk1]. We outline the general method and then demonstrate it with several types of computational workflows: classification, latent semantic analysis (LSA), quantile regression (QR), random data generation (RDG) and recommendations. We show how to leverage the engine by bringing your own templates (BYOT). The demonstrated functionalities use the packages in the project ""NLP Template Engine"" [AAr1], which is based on FindTextualAnswer [WRI1, JL1]. We concentrate on data science (DS) and machine learning (ML) computations, but the described method and overall algorithm are applicable to computation workflows from other fields.

Using the Dataset Query Language to Query Relational Databases

Leonid Shifrin Download Presentation Watch Video

Dataset has become a widely used tool to process and manipulate data in the Wolfram Language, but its scope has been limited to the data stored in kernel's memory. In this talk, I will discuss the general strategy to extend the scope of Dataset's query language to query other types of data sources and illustrate general ideas with examples of using Dataset's query language to query a relational database (SQLite).

Modeling X with PDEs

Oliver Ruebenkoenig Download Presentation Watch Video

In this talk, we will cover the usage of the NDSolve family of functions and the finite element method. The aim is to enable attendees to set up partial differential equation models from various fields of engineering. The focus will be on recent development such as axisymmetric regions and hyperelastic solid mechanics.

Interactive Analysis of Complex Optical Phenomenon

Jonathan Heinz Download Presentation Watch Video

Since 2016, Airy Optics Inc. has leveraged interactive Mathematica features for the real-time analysis of cutting-edge imaging systems, such as design of optics for augmented and virtual reality. Losses of image quality that are based in polarization are often poorly understood, and can be the result of the delicate interaction of many elements within an optical system. Airy Optics' premier software Polaris-M was written in Mathematica as the first dedicated tool for analyzing these effects using real-world data and scientifically proven modeling. This presentation will examine functions and displays written for Polaris-M and for broader demonstrations that leverage the power of Mathematica's interactive features to present complex concepts and data in a simple and easy-to-understand manner. One such function expands the built-in Grid function to better format and display multifaceted datasets, both improving readability and ease of data interaction. Another uses a custom-built framework to allow easy styling control of composed plots of many surfaces. A third example uses dynamic interactivity to condense a variety of analysis functions for high-dimensional measured polarization data into one flexible display. Airy Optics also uses simple, interactive Manipulate examples to teach optical engineering students basic polarization concepts, including helping them visualize polarization states in terms of ellipticity and the degree of polarization. All of these tools have helped us provide faster analysis to our customers in visually engaging ways, contributing substantially to the development of the next generation of AR/VR headsets as well as many other exciting applications.

Office Hour: Machine Learning

Giulio Alessandrini, Timothee Verdier, Jesse Galef

Drop in and chat with our machine learning experts. Ask questions, get help, or just say hello!

Workshop: Hands-on Paclet Development

Todd Gayley, Bob Sandheinrich, Bradley Ashby, Connor Gray, Richard Hennigan, Brian Van Vertloo Watch Video

Modeling Chemical Reaction Networks with Token-Event Graphs

Robert Nachbar Download Presentation Watch Video

The recent addition of PatternReaction and ApplyReaction to the Wolfram Language now allows the user to programmatically manipulate the atoms and bonds in Molecule objects in a chemically meaningful way. This talk will show how these functions can be used with token-event graphs to model chemical reaction networks. Examples with the Diels–Alder reaction, Cope rearrangement, Wagner–Meerwein rearrangement and terpene biosynthesis will be demonstrated. As these networks represent multiway systems, they can also be studied from that perspective with branchial and causal graphs.

Multi-framework Neural Networks

Matteo SalvarezzaDownload Presentation Watch Video

In this talk I will present NetExternalObject, a new symbol in Wolfram Language 13.2. It exposes new functionality to interface with external deep learning frameworks and run neural networks in their native format without having to manually install any external software.

Office Hour: Numerical Computing

Rob Knapp, Paritosh Mokhasi, Nina Dokeva, Oliver Ruebenkoenig

Drop in and chat with our experts. Ask questions, get help, or just say hello!

What's New in Calculus & Algebra

Devendra Kapadia Download Presentation Watch Video

I will give an overview of upcoming features related to calculus and algebra in the Wolfram Language. These features include dramatic performance improvements in polynomial algebra functions and in linear algebra for matrices of polynomials, new NFractionalD and NCaputoD functions for numerical computation of fractional derivatives, algorithmic advances for symbolic and asymptotic solutions of differential equations, large-scale table lookup for Laplace transforms, support for computing power series representations of functions using Asymptotic, and new built-in structured matrices for efficient computations in numerical linear algebra. I will also discuss current projects related to documentation, engineering education and online courses in this area.

High-Performance Polynomial Algebra

Adam Strzebonski Download Presentation Watch Video

The upcoming release of Mathematica includes significant performance improvements in polynomial algebra functions and in linear algebra for matrices of univariate polynomials. The release also includes functionality extensions in polynomial algebra over Zp The improvements and extensions are based on the FLINT library. In my talk, I will discuss the improvements and show some examples and performance comparison results.

Machine Learning Automation: The Next Generation

Timothee Verdier Download Presentation Watch Video

Introduction to a Unified Model of Cellular Automata

Rodrigo Obando Download Presentation Watch Video

This is an introduction to a unified model of Cellular Automata in which a rule is represented not by a single function but by a vector of functions we call genes. These functions can be ordered so that they maintain the same order regardless of the rule space where they are realized. This order allows the vectors to be normalized and positions all rules in a unit hypercube. There are operations on the rules that are simple to implement in the normalized space, such as Modulo-Additive Equivalent rules, known as complement in the binary k-value case. We also introduce the concept of symmetry index based on the number of symmetric genes that make up the rule. We also show the concept of canonical rule which is the rule number for the minimum value of r where the rule can be realized. There are many other functions that are introduced, such as genetic trees, equation trees, evolution trees and canonical trees, among others. There are also functions to perform reflection, reversibility, monotonicity and symmetry tests. Also, measures such as Langton's Lambda, Delta parameter and symmetry index are implemented.

Introduction to Fractional Calculus

Tigran Ishkhanyan Download Presentation Watch Video

Fractional calculus develops the theory of differentiation and integration of any real or complex order. It extends the basic operations of classical calculus to fractional orders and studies the methods of solving differential equations involving these fractional-order derivatives and integrals. Nowadays fractional calculus is gaining more and more interest as it is able to generalize the well-known integrodifferential equations to ones containing fractional order derivatives or integrals of the target function—these more general equations are able to precisely describe different real-world phenomena. This talk will give a brief introduction to fractional calculus, an overview of the FractionalD and CaputoD functions standing for the two main definitions of fractional order derivatives, their numerical analogues NFractionalD/NCaputoD functions, learning resources for this subject and some real-world applications of this theory.

Train a Production-Level Image Classifier

Maria Sargsyan, Jofre Espigule-Pons Download Presentation Watch Video

Office Hour: Graphs & Networks

Yan Zhuang, Charles Pooh

Drop in and chat with our experts. Ask questions, get help, or just say hello!

Supreme Court Justices Move Left Over Time (More than They Move Right)

Mark Fisher, Nicholas Georgakopoulos Download Presentation Watch Video

Martin and Quinn (2002) apply Item Response Theory (IRT) to Supreme Court votes in order to infer the "location" of justices on a simple left/right spectrum. We modify their model, adopting a non-conjugate prior for the case parameters and allowing justices' locations to vary with each change in the court's composition (instead of with each term).

We examine the joint distribution of justices' locations in their first and last courts, showing that the tendency to move left over the course of their tenures is greater than the tendency to move right. Although this finding is not new, we provide interesting illustrations via conditional distributions.

The model is simple. The only inputs are the justices' votes: vjk=1 indicates that justice j voted with the majority on case k and vjk=0 indicates a vote with the minority. Let vjk∼Bernoulli(πjkt), where πjkt=Φ((θjt–γk)/δk), Φ(·) is the CDF of the standard normal distribution, θjt is the location of justice j at "time" t, and (γkk) are the case parameters.

The posterior distribution is computed using a Gibbs sampler. We use a Metropolis step to draw the case parameters conditional on the locations. We discretize the location space and adopt a hidden Markov model wherein the locations become the hidden states. We use the forward-backward algorithm to draw the locations conditional on the case parameters. Joint work with Nicholas Georgakopoulos.

Progress in Symbolic Differential Equations

Aram Manaselyan Download Presentation Watch Video

During this talk I will give an overview of recent developments, new features and improvements in Wolfram Language related to symbolic solutions of ordinary differential equations. I will begin by speaking about the recently introduced DSolve option IncludeSingularSolutions, which allows one to get the singular solutions of a nonlinear ODE together with the general solution. Next, the significant improvements in the method of undetermined coefficients will be discussed, which allows one to find fast and simple solutions for inhomogeneous linear ODEs with constant coefficients. Finally, in the last part of the talk I will speak about a new type of ODE: fractional differential equations. Fractional differential equations are differential equations involving fractional order derivatives. These are generalizations of the ordinary differential equations that have attracted much attention and have been widely used in engineering, physics, chemistry, biology and other fields. I will demonstrate how to solve some linear fractional ODEs in Wolfram Language using DSolve and AsymptoticDSolveValue. I will end by discussing the updated tutorial on symbolic differential equation solving using DSolve.

Data Science: Processing Relational Data for Business or Research

Daniel Carvalho Download Presentation Watch Video

From small to big companies, operations work on top of traditional relational databases (RDBMS) and SQL is a lingua franca in business. To keep track of complex activities with vast amounts of transactions, lots of data is produced with ERPs, CRMs, BI, financial systems, RH, sales and internally built custom enterprise systems too. We can explore this information better to find patterns and insights for companies. The Wolfram Language can be used to import, process and combine all this info and relate it to real-world curated data. This data science hype brings P&D to many companies that can add value to the business and avoid risks.

Enabling Mathematical Search inside Wolfram Mathematica

Flavio B. Gonzaga Download Presentation Watch Video

SearchOnMath is a search engine specializing in formulas and text from several domains with math content. The web version indexes domains like Wolfram MathWorld, Stack Exchange (Math, Physics, and MathOverflow), Wikipedia, and others. The scientific version indexes preprints from arXiv, allowing users to submit queries and filter results by year, author, and/or area of knowledge. The SearchOnMath core is mainly composed of two parts: (i) Lexical analysis, which makes representation independent of notation. For example, variables can be denoted by Roman letters (x, y, z, x0, x1, …), Greek letters (α, β, …), and so on. (ii) Degree of similarity analysis, using a powerful engine to compare the structure of the target formula with that of all other formulas in the database. Formulas that are more similar in structure tend to appear first on the results page. This presentation shows a detailed view of SearchOnMath, including the ability to search arXiv preprints. It also shows how to integrate the SearchOnMath API into Wolfram Mathematica, including access to all features available on the SearchOnMath website. This integration enables countless possibilities, such as instantly searching from inside a Mathematica notebook all the main websites of mathematical content as well as all of arXiv.

Changing Variables and Other Calculus Tools

Itai Seggev Download Presentation Watch Video

In recent versions, we have added functionality to make Wolfram Language more directly applicable to the calculus classroom. I will focus, in particular, on the new functions ImplicitD, IntegrateChangeVariables and DSolveChangeVariables. Other related enhancements will be covered as time permits.

Logistic Regression and MLR for Controlling Shape Problem during Hot Strip Milling

Guarav Vishal Download Presentation

The dual-phase low-Cr ferritic stainless steels are characterized by severe poor shape, especially at the outer diameter (20–30%) of the hot pickling line (HPL) output coil or inner diameter of the hot strip mill (HSM) output black coil. The objective of this work is to study the parameters affecting the waviness level and affected length. We will be using the various statistical capabilities of Wolfram Language, including ANOVA, multiple linear regression and logistic regression, to study the parameters affecting the shape of the coil during the entire rolling operation (including the chemistry of the grade). Overall, we will explore the ways in which Wolfram Language can be used to solve industrial challenges.

Office Hour: Calculus & Algebra

Devendra Kapadia, Adam Strzebonski, Aram Manaselyan, Tigran Ishkhanyan
The calculus and algebra team is available to answer your questions and offer assistance. Drop in and say hello!

The Question Framework: From Assessments to Applications

Bob Sandheinrich Download Presentation Watch Video

See how to create courseware using the question and assessment framework. This talk shows how each piece of the system works, from custom, automated assessment functions to full quizzes.

Visualization Overview

Brett Champion Download Presentation Watch Video

Five Tricks to Highly Effective Visualizations

Tim Shedelbower Download Presentation Watch Video

One of many wonderful components waiting to be discovered in Wolfram Language is powerful graphics. I will share some of my techniques developed to harness this power for documentation illustrations and marketing images. Learn style, lighting and camera tips to solve mysteries you may have encountered while creating your visualizations. Leave encouraged to explore the broad range of graphic functions.

Unity 3D Math Sims with Companion Wolfram Notebooks for Active Learning

Jay Alan Jackson Download Presentation

This talk will present four Unity 3D Math Games/Simulations with companion Wolfram Notebooks, the development of which is ongoing. The potential for melds such as these to evolve into active learning resources for science education will be explored.

Multipanel Visualization

Brett Champion Download Presentation Watch Video

Digital Fluid Mechanics Laboratory: Sudden Expansion Pressure Loss

August Young (Frechette) Download Presentation Watch Video

This work details the assignment of a digital, fluid mechanics laboratory to a class of undergraduate engineering students at Duke University. In this laboratory, students are asked to numerically simulate flow through a sudden expansion. The primary learning objective of this laboratory is for students to understand the physical limitations associated with this experiment. Physical measurements disrupt the flow through such a geometry, imparting needless—and simultaneously unmeasurable—sources of error into the measurement process. The digital nature of this laboratory allows students to take these measurements without disturbing the flow, altogether avoiding this issue. The second learning objective of this laboratory is for students to gain experience utilizing the Wolfram Language—a severely underutilized tool in engineering curricula. Among the many advantages to utilizing Mathematica for this laboratory exercise is the fact that students can observe and understand the mechanics of the numerical flow solver they are tasked with using, which is often not the case; commercial flow solvers operate more like "black-box" technologies and low-level programming languages require advanced knowledge to code such simulations. In this work, we also document the experience of the students as they complete the laboratory assignment. The novelty of this work is attributed to the fact that no such exercise has been previously detailed in the literature. And while this work describes a fluid mechanics laboratory, the digital nature of this exercise can easily be implemented in any course to supplement student learning when physical experimentation is challenged by nuances like the observer effect.

Scales & Guides—How to Visualize Anything

MinHsuan Peng Download Presentation Watch Video