The impact of differentiable programming: how ∂P is enabling new science in Julia
Fully incorporating differentiable programming (∂P) into the Julia language has enabled composability between modern machine learning techniques and existing high performance computing (HPC) modeling and simulation without sacrificing expressivity. Most notably, this has meant that small neural networks can be embedded within larger models whose other behaviors are fully understood and can be concretely represented. Smaller neural networks, in turn, are easier to train and interpret. It has also enabled complex computations to be embedded within cost functions for fast and robust reinforcement learning. In this talk, we’ll walk through several concrete examples and demonstrate how the combination of ∂P with Julia’s generic programming has enabled powerful and expressive new models.
Presented by Matt Bauman
Видео The impact of differentiable programming: how ∂P is enabling new science in Julia канала ACM SIGPLAN
Presented by Matt Bauman
Видео The impact of differentiable programming: how ∂P is enabling new science in Julia канала ACM SIGPLAN
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
Differentiable Programming with Julia by Mike InnesAugmenting Differentiable Simulators with Neural Networks to Close the Sim2Real GapTesla Valve | The complete physicsBut what is a neural network? | Chapter 1, Deep learningMarvin Minsky - The beauty of the Lisp language (44/151)Models as Code: Differentiable Programming with ZygoteAnima Anandkumar (CalTech): "Infusing Physics and Structure into Machine Learning"DiffTaichi: Differentiable Programming for Physical SimulationMIA: John Ingraham, Learning protein structure with a differentiable simulatorJuliaCon 2020 | Doing Scientific Machine Learning (SciML) With JuliaDifferentiable Programming in C++ - Vassil Vassilev & William Moses - CppCon 202110 Different Programming Languages and Their UsesBoeing Colloquium: Julia: Differentiable Programming and Software 2.0JuliaCon 2019 | Differentiate All The Things! | Mike InnesLarry Wall: 5 Programming Languages Everyone Should Know | Big ThinkI Built an A.I. Voice Assistant using PyTorch - part 1, Wake Word DetectionThe state of JuMP | Oscar Dowson | JuliaCon2021DDPS | Differentiable Physics Simulations for Deep LearningPhysics-based differentiable rendering (CVPR 2021 tutorial)Tobias Pfaff (DeepMind): Learning to Simulate Complex Physics with Graph Networks