Wolfram Screencast & Video Gallery

This page requires that JavaScript be enabled in your browser.
Learn how »

User-Extensible Layers for Neural Networks

Matteo Salvarezza presents a new neural net layer in Wolfram Language 12.2. Leveraging the new Wolfram Compiler, it allows users to implement custom operations and freely use them with the rest of the framework, including training functionality.

Was this video helpful?

Channels: Technology Conference

SORT BY: Latest | A-Z
1311 videos match your search.
User-Extensible Layers for Neural Networks30:05
Multi-language Data-Wrangling Conversational Agent30:41
Quick, Fun, Hack Side Projects with the Wolfram Language and Cloud24:26
Human in the Loop: Interpretable Machine Learning30:20
Ludic Uses of Technology for Science and Mathematics Education27:53
Physical Simulation in the Wolfram Language18:56
Neural Net in your Phone: From Training to Deployment through ONNX26:45
Text and Image Mining for Historical Research36:24
Dynamic Template Boxes and Other Typesetting Advances38:09
Geodata Applied to Civil Engineering: The LEP/LHC Tunnel at CERN32:04
Interactive Math Exercises in MyLab25:23
State of the Notebook Front End34:07
What's New in Cryptography28:52
Graphics: Textures, Shading and Other Enhancements27:56
What's New in Graphs and Networks25:09
Graphics: New and Experimental Features37:07
Wolfram Eureka! Program—Discussion with a Startup Founder Philip Maymin57:54
Building Contextual Interfaces35:55
Computational Models in Human-Scale Problems: COVID Costa Rica29:38
Connecting to the Wolfram Computational Geometry Engine20:50