2023

Márcio Rosa

Professor of Mathematics, IMECC-UNICAMP

Areas: Education, Mathematics, Mathematics Courseware Design

For 20 years, Márcio Rosa has been making pedagogical innovations on the principle that university students should be using software, including the Wolfram Cloud, to continue their education in higher mathematics. He believes students should be trained to use software as a tool to aid their endeavors rather than learning to replicate the software’s functions. The geometric approach is reinforced so that the student, when studying and solving problems, is able to produce images with software and interpret them. Rosa has published various articles and supervised theses based on his experience and unique approach to mathematics education.

2023

Patrick Scheibe

Research Scientist, Max Planck Institute for Human Cognitive and Brain Sciences

Areas: Data Analytics, Programming, Software Engineering

Patrick Scheibe boasts a dynamic and illustrious career journey in academia and industry. He spent over a decade at Leipzig University, where he played a pivotal role in leading an image and data processing unit, enabling researchers to quantify medical and biological experiments easily. During his PhD studies, he took a deep dive into the intricacies of the human fovea, extensively utilizing Wolfram Language to model and quantify this crucial eye region from optical coherence tomography scans. Subsequently, Patrick’s expertise took him to the neurophysics department at the Max Planck Institute for Human Cognitive and Brain Sciences, where he continues to work on data processing for quantitative MRIs.

Patrick is a highly versatile professional with a wide range of expertise beyond academia. He has worked as a consultant using Mathematica on various projects focused on simulations, modeling and data analyses in diverse domains for companies like Daimler, Procter & Gamble and Dow Chemical. Patrick has been developing and maintaining the Wolfram Language integration for JetBrains IDEs since 2012. His exceptional skills and expertise have led him to join the IntelliJ Platform SDK team at JetBrains. In addition, Patrick has developed several syntax highlighters for Wolfram Language, one of which has been used on the official Mathematica Stack Exchange site, where he is an enthusiastic moderator and member.

2023

Esma Gel

Cynthia Hardin Milligan Chair of Business and Professor of Supply Chain Management and Analytics, College of Business, University of Nebraska–Lincoln

Areas: Modeling Dynamical Systems with Mathematica, Research and Analysis

In her previous role as an associate professor in the School of Computing, Informatics, and Decision Systems Engineering at Arizona State University, Esma Gel used Mathematica for a system dynamics model related to the spread of COVID-19.

Gel’s team, ASU METAz, helped guide the Arizona Department of Health Services by supplying predicted outcomes to various “what if?” policy questions. The team periodically released accurate projections for cases, hospitalizations and deaths in Arizona for more than 15 months, often being featured in mainstream media outlets.

2023

Martijn Froeling

Assistant Professor, University Medical Center Utrecht

Areas: Image Processing, Research and Analysis, Software Development

Martijn Froeling is an assistant professor specializing in quantitative neuromuscular magnetic resonance imaging (MRI) at the University Medical Center Utrecht. His work revolves around enhancing MRI techniques to better understand muscle function and diseases.

MRI scans provide valuable data, but they need careful processing and analysis. That’s where Froeling’s QMRITools paclet comes in. The paclet is a handy toolkit for experimental design, data analysis and teaching. Since its launch in 2012, it has been used in over 50 scientific papers. Originally created to analyze muscle diffusion-weighted imaging data, QMRITools has expanded its scope. It now includes features like cardiac analysis (including tagging and T1 mapping), Dixon reconstruction, EPG modeling and fitting, J-coupling simulations and more.

The paclet currently offers over 450 custom functions, making it a valuable resource for researchers. Plus, there’s extensive documentation with more than 750 pages, and each toolbox comes with demonstrations. With these tools, Froeling aims to simplify quantitative MRI analysis, benefiting our understanding of muscle injury and disease.

2023

Oliver Knill

Preceptor and Digital Media Specialist, Harvard University

Areas: Computational Thinking, Education, Geometry, Mathematics

Mathematica is vital to Oliver Knill’s teaching and research. In teaching, it produces professional graphics for handouts, facilitates visualizations and animations, and serves as a platform for innovative student projects. It’s also essential for vetting assignments and examples, enabling a quick search for appropriate problems for both homework and exams. Knill has employed it to design 3D printable objects, generate high-resolution animations and illustrate musical concepts like Markov chain–generated music.

In his research, Knill’s primary laboratory is Mathematica. Currently, he is delving deeper into discrete geometry, probability, spectral theory and linear algebra. He is thrilled about uncovering previously undiscovered relationships and enhancing proofs with code. This allows any curious individual to explore the underlying structure. Mathematica code is close to natural language, acting as a runnable pseudocode. While examples can elucidate a theorem, providing code that showcases it using random structures is not only thrilling but also validates the result’s efficacy.

2023

Picket Pharmaceuticals, Inc.

Accepted by: Joshua Kriger and Lauren Williams

Areas: Data Analysis, Data Analytics, Economic Research and Analysis

The foundation of Picket Pharmaceuticals, Inc.’s approach is to first acquire, then integrate, large healthcare datasets—such as shortage data, manufacturing information, unit usage, pricing, price variation and many more—that capture the universe of healthcare interactions that surround each patient’s walk from diagnosis to completion of care through their piece of the healthcare system.

Using the Wolfram technology stack, Picket has conducted interesting work with insights on the points of failure and where inefficient markets exist in the supply chain. Of note, computation techniques used include projecting large amounts of healthcare supply and medication usage data into images. These images become the data fed to repurposed visual neural net training procedures that result in AI/machine learning models that are able to accurately recognize signals that predict future drug shortages.

Working with Wolfram’s Consulting Group, Picket has also verified a derived new class of economic measures, titled the Sutherland measures, made feasible by taking into account special economic qualities and situations of supplied medicines for the generic drug markets.

2023

Sander Huisman

Professor, Physics of Fluids, University of Twente

Areas: Data Analysis, Physics, Software Development, Software Engineering

Sander Huisman has been using Mathematica since 2003 for the processing of all his data, creating figures and visualizations and doing complicated fits and optimizations. Furthermore, he uses Mathematica’s interactive capabilities to generate illustrative examples in his fluid mechanics classes. He also uses it recreationally for the production of generated art for the yearly GENUARY event. He is also a contributor to the Wolfram Function Repository, having created over one hundred functions.

2023

Mark Rawlins

Executive Chairperson and Chief Engineer, Energy and Combustion Services

Areas: Mechanical Engineering, Research and Analysis, Software Engineering

Energy and Combustion Services offers global energy management analytics and autonomous measurement systems for large-scale mining and industrial manufacturing. Mark Rawlins is a professional engineer (mechanical), certified energy manager, and measurement and verification professional. He specializes in energy system modeling for efficiency and productivity, using digital twins to simulate and support new designs. His primary goal is aiding companies in transitioning to net-zero carbon emissions while maintaining efficiency. He also develops advanced metering systems that provide insights into energy and process deviations, some operating autonomously.

Wolfram Language is foundational to his R&D work, which includes a road condition monitoring system that marries vision-based road defect detection and location with vehicle dynamics and vibration signal processing using edge computation to report road and safety conditions autonomously. Separate devices can communicate and accept instructions from each other for extended inspections.

2022

William A. Sethares

Professor, Electrical and Computer Engineering, University of Wisconsin–Madison

Areas: Computational Humanities, Computational Thinking, Computer-Aided Education, Courseware Development, Engineering, Image and Signal Processing, Image Processing, Signal Processing

Bill Sethares is a researcher and professor of electrical and computer engineering at the College of Engineering at the University of Wisconsin–Madison, focusing on signal processing with applications in acoustics, image processing, communications and optimization.

At the University of Wisconsin–Madison, Sethares attracts students from majors beyond engineering with his computationally rich image processing course material and project-based learning (all Wolfram Language–based, of course!). Sethares is a founding member of the LEOcode project and brings computation to art historians in the form of applications used to find patterns in watermarks and canvases. These can help to identify and date historical papers and paintings.

2023

Tyson Jones and Simon Benjamin

Tyson Jones, Postdoctoral Researcher, University of Oxford
Simon Benjamin, Professor of Quantum Technologies, University of Oxford

Areas: Physics, Programming, Software Development, Software Engineering

Tyson Jones is a postdoctoral researcher at the University of Oxford, studying first-generation quantum computers and their simulation via high-performance classical computing in the areas of quantum computing, high-performance computing, scientific simulation and software development. He is also a senior quantum software engineer at Quantum Motion Technologies and a consultant for the UK’s National Quantum Computing Centre.

Jones’s doctoral work included the creation of QuESTlink, an open-source WSTP-powered package for simulating quantum computers, integrating the QuEST project’s hardware-accelerated numerics with Mathematica’s powerful symbolic engine. QuESTlink combines a plethora of Wolfram facilities, novel algorithms and high-performance computing techniques behind an intuitive API, enabling research-frontier computation through only a few lines of code.

Simon Benjamin, principal investigator (PI), is a professor of quantum technologies with the Materials Department at the University of Oxford. He leads a group of 17 applied theorists who look at diverse aspects of quantum computing, including architectures, fault tolerance and algorithms that are robust against hardware imperfections. His team created QuEST, a world-leading tool for classical emulation of quantum devices.

2023

Sandipan Bandyopadhyay

Associate Professor, IIT Madras

Areas: Computational Thinking, Education, Physics

Sandipan Bandyopadhyay is an educator and researcher in the fields of mechanisms and robotics. He specializes in theoretical and computational kinematics, in particular in the domain of spatial parallel manipulators, such as the Stewart platform.

Bandyopadhyay’s research involves highly demanding symbolic computations, for which he finds a trusted partner in Mathematica. In at least 20 of his journal publications, the symbolic capabilities of Mathematica have played a significant role. Moreover, the flexibility of Wolfram Language has allowed him to develop algorithms and modules to explore deeper into algebraic geometry and kinematics and create customized tools for analyzing problems using hyper-complex numbers, such as dual numbers and dual quaternions. He uses the dynamic visualization capabilities of Mathematica to bring virtual robots to life, enabling his students to manipulate them and develop a better understanding of complicated motions of constrained multibody systems.

2023

Australian Christian College

Accepted by: Jeremy Kwok, Director of Technology

Areas: Computer-Aided Education, Education

Australian Christian College (ACC) is Australia’s largest non-government distance education provider, with four schools equipped to provide a hybrid of in-house and fully remote learning. Covering K–12 education, its mission is to have all students flourish to their full potential and be a positive influence on the world. Rather than rely on traditional publishers, ACC wanted more consistent content for online and in-person assessment that showcases the ingenuity of their instructors.

ACC is now rolling out a Wolfram eLearning environment integration in their current Canvas environment. This integration will initially be used for assessments with 1,500 high-school students, which will quickly grow to 3,000 students and beyond as they implement the system for grades 7–12. With ACC’s announcement of a new STEM-focused campus in Western Sydney, this automated assessment system will be even more important in the future as enrollment continues to grow.

2023

J. William Helton

Professor Emeritus of Mathematics, University of California San Diego

Areas: Geometry, Mathematics, Software Engineering

J. William Helton’s group developed the package NCAlgebra for doing general-purpose noncommutative algebra in Mathematica. It began around 1990 and has been extended continually since then. From his work at the origins of noncommutative geometry and H-infinity control, Helton kept seeing such noncommutative formulas and hoping computer algebra could help. So, with Bob Miller, he started NCAlgebra and developed algorithms to find out. Around the year 2000, linear control theory shifted away from equalities to inequalities, e.g. from Riccati equations to linear matrix inequalities. This motivated Helton and a few others to begin what has developed into an elegant theory of noncommutative inequalities, to wit, a noncommutative version of real algebraic geometry. NCAlgebra seriously accelerated (and was accelerated by) this development.

A booming area full of noncommutative algebra is quantum information theory, and that is the main direction of current NCAlgebra development. Major contributions to NCAlgebra are being made by Mauricio de Oliveira and have also come from Mark Stankus and from many University of California San Diego students.

2025

WebAssign Engineering Team

Cengage Group, Inc

Areas: Education, Mathematics Courseware Design

WebAssign began in 1997 at North Carolina State University as an early online homework platform for physics students. Over time, the technology grew into its own company, eventually providing online homework for millions of students globally.

In 2008, WebAssign took a major step forward by adding symbolic math to its homework system and chose Wolfram as its technology partner. Wolfram’s computational engine enabled real-time evaluation of student answers for accurate grading.

After Cengage acquired WebAssign in 2016, Wolfram technology became deeply integrated across a wide range of textbooks. Today, approximately 1.1 million WebAssign questions use Wolfram Language for instant grading and intelligent feedback. These textbooks span areas like algebra, pre-calculus, calculus, differential equations and statistics.

2025

Aninda Sinha

Indian Institute of Science, India
University of Calgary, Canada

Areas: Cosmology, Mathematical Physics, Quantum Field Theory, String Theory

Aninda Sinha is a theoretical high-energy physicist. His research focuses on quantum field theory, string theory, cosmology and mathematical physics.

Sinha’s research is both numerical and analytical in nature and makes heavy use of the capabilities of Mathematica. He has used Mathematica as a “theoretical experimentalist” in many of his 80 publications. His research on the bootstrap makes heavy use of the symbolic manipulation capabilities as well as the high-precision numerical capabilities of Mathematica. Recently, he found an infinite number of new formulas for pi in a paper that was published in Physical Review Letters, which was selected by Scientific American as one of the seven coolest mathematical discoveries of 2024. Mathematica provided a versatile platform to verify these results and to gain fresh, new perspectives on string theory.

2025

Gareth Russell

New Jersey Institute of Technology

Areas: Conservation Biology, Ecology

Gareth Russell is a computational ecologist and conservation biologist, focusing on spatial and movement ecology. He started using Mathematica at Version 4, mainly for population modeling, and expanded his use as its capabilities have grown. Currently he uses it to build and fit statistical models of animal movement in complex real-world landscapes, which also requires the processing of geospatial movement tracks and the processing and sampling of remotely sensed data layers.

From the beginning, Gareth also brought Mathematica into his teaching, from complete graduate courses in computational ecology and statistics to course modules on epidemiology to general education labs that introduce freshmen to chaos, fractals and other amazing phenomena. For a while, he ran a self-hosted webMathematica site that provided free tools for doing common basic calculations in conservation biology. Recently, he has been experimenting with building specialized CNNs for separation of certain hard-to-distinguish species encountered during biodiversity surveys.

2025

Martin Ricker

Instituto de Biología Universidad Nacional Autónoma de México (UNAM)

Areas: Applied Statistics, Forest Science

Martin Ricker studied biology at the University of Würzburg in Germany, followed by his doctoral research at Yale University’s School of Forestry and Environmental Science. Since 1995, he has worked as a research fellow at the Institute of Biology of the National Autonomous University of Mexico in Mexico City to develop methods and knowledge for management and conservation of species-rich tropical forests. He has been an enthusiastic user of Mathematica since 1991, when he purchased the software for the first time, for his doctoral studies, using it for analyses of databases, generation of graphs, and regression and statistical analyses, as well as explorations with calculus. Over the years, a number of Mathematica notebooks have been published together with scientific articles about topics as varied as solving linear regression without skewness of the residuals’ distribution, statistical age determination of tree rings and a generalization of the exponential function to model growth. Currently, he is working with Mathematica on articles about generalized multivariate analysis of variance (GMANOVA), the numerical calculation of the inverse of the exponential integral Ei(x) and modeling tree growth curves indirectly with piecewise linear regression when tree ages are unknown.

2025

Nathan Myhrvold

Intellectual Ventures

Areas: 3D Printing, Asteroids, Biostatistics, Computer Science, Global Health and Development, Image and Signal Processing, Image Processing, Nuclear Energy, Paleontology, Planetary Science, Statistics, Theoretical Physics

Dr. Nathan Myhrvold is a prominent scientist, technologist, inventor, author and food photographer. In a 14-year tenure at Microsoft, Myhrvold led advanced technology and business development groups, founded Microsoft Research, managed an R&D budget of $2 billion and served as the company’s chief strategist and chief technology officer. Myhrvold previously had cofounded Dynamical Systems Research, a software company, and worked as a postdoctoral fellow in the Department of Applied Mathematics and Theoretical Physics at Cambridge University, where his research with Professor Stephen Hawking centered on quantum theories of gravitation. Myhrvold has published peer-reviewed research in planetary science, paleontology and climate science.

In 2000, after retiring from Microsoft, Myhrvold founded Intellectual Ventures (IV), which he leads as CEO and one of its most prolific inventors, with more than 850 US patents awarded. Myhrvold is vice chairman of TerraPower, a pioneer in advanced nuclear energy. He also has served as an adviser to startup companies that IV has launched to commercialize advanced metamaterials technologies.

Myhrvold is the author of seven Modernist Cuisine culinary and photography books and a pioneer of innovative photography techniques and equipment. His most recent landscape photography book Natural Wonders is published by National Geographic and is on sale now.

Myhrvold holds a doctorate in theoretical and mathematical physics, as well as a master’s degree in mathematical economics, from Princeton University. His master’s degree in geophysics and space physics, as well as his bachelor’s degree in mathematics, are from the University of California, Los Angeles.

2025

Thomas Landsberger

ENODA Ltd

Areas: Control Systems, Economics, Electrical Engineering, Electromagnetic, Energy and Sustainability, Software

Thomas Landsberger, Dipl.Ing., MSc, is an engineering leader whose innovative use of Wolfram Language has reshaped how complex control systems are modeled and implemented at ENODA. With a background in electrical and control engineering; over 20 years of software experience, including at Microsoft; and an additional degree in economics, Thomas brings a rare interdisciplinary perspective to solving high-impact technical challenges.

At ENODA, Thomas played a key role in developing the ENODA PRIME® Exchanger, a first-of-its-kind dynamic power-flow technology designed to replace traditional grid transformers, increase distribution capacity and enable large-scale decentralized renewable generation. Faced with the complexity of accurately modeling and controlling the system’s 12-coil, 12-degree-of-freedom architecture, he used Wolfram Language to derive control algorithms from first principles and build a code-generation workflow that ensures faithful, verifiable firmware implementation on real-time hardware, avoiding the risks of manual equation handling and translation to C.

Originally adopting Wolfram Language for advanced computational geometry problems in 3D-printing software development, Thomas has since become a driving force in expanding its use across ENODA. He has promoted Wolfram-based methods in research spanning physics, electrical engineering and signal processing, and organized training to embed Wolfram capabilities within the wider technical team. His ongoing work includes streamlining Wolfram Language-to-C/C++ deployment for real-time systems and developing tools that integrate Wolfram Language models with MATLAB/Simulink circuit simulations.

2025

Gabriel Landi

Associate Professor, Department of Physics and Astronomy, University of Rochester, NY

Areas: Energy Harvesting Devices, Open Quantum Systems, Quantum Computing, Quantum Sensing, Quantum Thermodynamics, Transport and Metrology

Gabriel T. Landi is an associate professor in the Department of Physics & Astronomy at the University of Rochester, where he directs the Quantum Thermodynamics and Quantum Transport group (QT2). He also serves as an associate editor of Physical Review Research. Professor Landi’s research sits at the crossroads of quantum information science, open quantum systems and nonequilibrium statistical physics. He is particularly recognized for his work in quantum thermodynamics and quantum transport: developing theoretical frameworks that reformulate thermodynamic laws in the quantum-coherent regime, analyzing quantum stochastic processes and investigating how coherence and fluctuations influence energy, entropy and information in small and strongly coupled quantum systems. His interests span theory of open quantum systems, quantum stochastic processes, quantum information theory and quantum metrology. A distinctive aspect of Professor Landi’s contributions is his design and deployment of a specialized computational framework built in Mathematica: the Melt library. Melt is a fully self-contained Mathematica notebook/package developed under his direction with the QT2 group, which provides users with a high-level yet transparent environment for simulating and analyzing open quantum systems, quantum information measures, Gaussian states, full-counting statistics and more. In his current role, he leads the QT2 research group, which applies advanced methods of quantum trajectories, full‐counting statistics and stochastic thermodynamics to explore fundamental questions (e.g. how irreversibility emerges at the quantum level) and to propose novel applications—including quantum sensing, thermal machines, energy harvesting at the nanoscale and quantum transport devices. Professor Landi’s work has gained widespread recognition in the community. His group publishes regularly in leading journals and is frequently invited to contribute to seminars and workshops on quantum thermodynamics and transport. He is also committed to mentoring the next generation of researchers and to advancing the theoretical foundations of quantum nonequilibrium physics.

2025

Thomas Hahn

Max-Planck-Institut für Physik

Areas: Computational Physics, Package Development, Physics, Quantum Field Theory

Thomas Hahn is a department leader at the Max Planck Institute for Physics. He is also on the executive committee for the Fachgruppe Computeralgebra (Subject Group in Computer Algebra). He is the author of numerous packages for performing calculations in quantum field theory, including FeynArts, one of the most highly cited Mathematica packages to date. He has applied his expertise to the study of high-energy computational physics and is personally the author of over one hundred scholarly works in this field.

2025

Mahn-Soo Choi

Professor of Physics, Department of Physics, Korea University
Director, School of Quantum, Korea University

Areas: Quantum Computing, Quantum Information, Theoretical Physics

Mahn-Soo Choi is a professor of physics at Korea University in South Korea and received a PhD in physics from POSTECH in 1998. Following a postdoctoral fellowship at the University of Basel in Switzerland and a research fellowship at the Korea Institute for Advanced Study, he joined the Faculty of Physics at Korea University in 2002.

Choi’s research interests have evolved from his doctoral dissertation in condensed matter theory to encompass quantum computation and quantum information. His areas of expertise include quantum algorithms, quantum simulations, superconducting qubits, spin qubits in quantum dots and circuit quantum electrodynamics in quantum hybrid systems, as well as mesoscopic transport.

In the early 2000s, Choi initiated the development of Mathematica packages specifically designed for students studying quantum many-body systems and quantum spin systems. His primary objective was to assist students in prioritizing essential physics concepts over technical calculations. These packages eventually evolved into Q3, a symbolic quantum simulation framework implemented in Wolfram Language. Q3 was released to the public in 2020 through a GitHub repository and was subsequently featured in his recent book published by Springer, titled A Quantum Computation Workbook.

2025

Sheldon Axler

Emeritus Professor of Mathematics, San Francisco State University

Areas: Education, Mathematics, Package Development, Publishing and Authoring

Sheldon Axler is professor emeritus at the San Francisco State University Department of Mathematics. He has served over 15 years in leadership roles as the Mathematics Department Chair and Dean of the College of Science and Engineering at San Francisco State University. He has also held faculty positions at Michigan State University, the Mathematical Sciences Research Institute at Berkeley, Indiana University and the Massachusetts Institute of Technology. He has authored numerous excellent texts on college and graduate mathematics, including Linear Algebra Done Right, Harmonic Function Theory, College Algebra and many others. He has authored a Mathematica package for harmonic function theory calculations as well as over a hundred scholarly works in mathematics.

2024

Thomas Wallek

Associate Professor, Graz University of Technology

Areas: Chemical and Process Engineering, Chemical Thermodynamics, Education

Thomas Wallek’s area of expertise is chemical thermodynamics and its application in chemical and process engineering, for which he uses Wolfram as an essential standard tool for both research and teaching. In his research, he focuses on thermodynamic modeling, the estimation of physical property data, the characterization of complex mixtures and molecular simulations.

In this context, his workgroup has been developing a Gibbs-ensemble Monte Carlo simulation program completely implemented in Wolfram, which is constantly being further developed and benefits from the diverse visualization and evaluation capabilities of Wolfram Language. In the context of teaching, Wallek is committed to the Technology Enhanced Learning (TEL) Marketplace initiative of Graz University of Technology that focuses on the development and scaling of TEL innovations for teaching. In particular, Wolfram Notebooks were created as the basis of a chemical thermodynamics course that was designed in the “inverted classroom” concept and enables students to acquire the material independently through self-study. For this purpose, manipulable diagrams and animated equations were integrated into the notebooks, with which students can interact and learn independently.

2024

David G. Stork

Stanford University

Areas: Computer Graphics and Visual Arts, Computer-Aided Education, Engineering, Image Processing, Machine Learning, Materials Science, Mathematics, Visualization

David G. Stork is an adjunct professor of electrical engineering, symbolic systems and material science and engineering, as well as an adjunct lecturer in computational mathematics and engineering at Stanford University, where he considers Mathematica to be a valuable teaching tool and resource. Here, he developed and teaches Computational Symbolic Mathematics, a Mathematica-based course for using computer algebra for solving difficult non-numerical mathematical problems. Stork is a graduate in physics from the Massachusetts Institute of Technology (MIT) and the University of Maryland. He has held faculty positions at Wellesley and Swarthmore Colleges; Clark, Boston and Stanford Universities; and the Technical University of Vienna. Stork has been a long-time friend of Wolfram, using Mathematica in teaching and research. He holds 64 US patents and has published over 220 scholarly papers and nine books and proceedings volumes, including Pattern Classification, Second Edition and Pixels & Paintings: Foundations of Computer-Assisted Connoisseurship.