Explore cutting-edge applications of deep learning and auto-differentiation techniques in molecular simulations through this comprehensive talk. Delve into active learning of machine learning potentials, deep neural network generative models for coarse-grained atomic system representations, and differentiable simulations for reaction path finding. Examine the challenges of overfitting and generalizability in AI for science, discussing the scalability of active learning in molecular simulations and the sensitivity of deep learning solutions to molecular problems. Gain insights into screening photoswitchable drugs, differentiable uncertainty, and the issues and tools needed for advancing this field.
Overview
Syllabus
- Intro
- Virtuous Cycle for Design
- Autodiff, Uncertainty, and ML Potentials
- Using Neural Network Potentials for Molecules
- Screening Photoswitchable Drugs
- Differentiable Uncertainty
- Beyond Forces
- Differentiable Simulations
- Issues and Tools Needed
- Q+A
Taught by
Valence Labs
Related Courses
-
End-to-End Learning and Auto-Differentiation - Forces, Uncertainties, Etc.
-
Differentiable Simulations for Enhanced Sampling of Rare Events
-
Training Neural Network Potentials: Bayesian and Simulation-based Approaches
-
Machine Learning Gradients in Molecular Simulations
-
Uncertainty-Aware Machine Learning Models of Many-Body Atomic Interactions
-
Machine Learning for Multi-Scale Molecular Simulation and Design
