Valence Labs Courses

Path Integral Stochastic Optimal Control for Sampling Transition Paths in Molecular Dynamics

Explore stochastic optimal control for sampling molecular transition paths, considering full geometry without relying on collective variables. Applicable to larger proteins like Polyproline and Chignolin.

• YouTube
• 1 hour 7 minutes
• On-Demand
• Free Video

Fragment-Based Hit Discovery via Unsupervised Learning of Fragment-Protein Complexes

Explore unsupervised machine learning for fragment-based hit discovery in drug development. Learn about FRESCO method, pharmacophore modeling, and its application in COVID-19 research.

• YouTube
• 54 minutes
• On-Demand
• Free Video

Calibration and Generalizability of Probabilistic Models on Low-Data Chemical Datasets

Explore calibration and generalizability of probabilistic models on small chemical datasets. Gain insights into model selection and feature choice for molecular property prediction and design.

• YouTube
• 53 minutes
• On-Demand
• Free Video

DGL 1.0 - Empowering Graph Machine Learning for Everyone

Explore Deep Graph Library 1.0's comprehensive support for Graph Machine Learning, featuring DGL Sparse for advanced models and scalability improvements for real-world applications.

• YouTube
• 1 hour 15 minutes
• On-Demand
• Free Video

Interpretable Chirality-Aware Graph Neural Networks for QSAR Modeling in Drug Discovery

Explore an innovative Graph Neural Network approach for drug discovery, incorporating molecular chirality and offering interpretable results to enhance quantitative structure-activity relationship modeling.

• YouTube
• 49 minutes
• On-Demand
• Free Video

Structure-Aware Protein Self-Supervised Learning

Explore structure-aware protein self-supervised learning, combining GNN models with protein language models to enhance structural information capture for improved downstream task performance.

• YouTube
• 30 minutes
• On-Demand
• Free Video

An Optimal Control Perspective on Diffusion-Based Generative Modeling

Explore optimal control theory's connection to diffusion-based generative modeling. Gain insights into stochastic differential equations, Hamilton-Jacobi-Bellman equations, and novel sampling methods for unnormalized densities.

• YouTube
• 1 hour 11 minutes
• On-Demand
• Free Video

Differentiable Simulations for Enhanced Sampling of Rare Events

Explore deep learning applications in molecular simulations, including surrogate functions, sampling complex distributions, and time-propagation of differential equations. Learn about active learning, generative models, and reaction path finding.

• YouTube
• 1 hour 1 minute
• On-Demand
• Free Video

Likelihood Training of Schrödinger Bridge Using Forward-Backward SDEs Theory

Explore likelihood training of Schrödinger Bridge using Forward-Backward SDEs theory for deep generative modeling, offering mathematical flexibility and optimization principles.

• YouTube
• 1 hour 8 minutes
• On-Demand
• Free Video

Geometric Clifford Algebra Networks and Clifford Neural Layers for PDE Modeling

Explore Clifford algebras in deep learning for PDE modeling, focusing on geometric transformations and fluid dynamics. Learn about Clifford neural layers, convolutions, and Fourier transforms to enhance neural PDE surrogates.

• YouTube
• 1 hour 32 minutes
• On-Demand
• Free Video

Neural Network Potentials for Low-Energy 3D Structure Generation and Reactivity Prediction

Explore neural network potentials for efficient 3D structure generation and reactivity prediction in molecular modeling, featuring Auto3D package and ANI-2xt model advancements.

• YouTube
• 50 minutes
• On-Demand
• Free Video

On the Expressive Power of Geometric Graph Neural Networks

Explore the expressive power of geometric graph neural networks, focusing on the Geometric Weisfeiler-Leman test and its implications for distinguishing geometric graphs while respecting physical symmetries.

• YouTube
• 1 hour 48 minutes
• On-Demand
• Free Video

Molecule Representation Learning: A Perspective from Topology, Geometry, and Textual Description

Explore molecule representation learning through topology, geometry, and textual description. Gain insights into AI-driven drug discovery and molecular modeling techniques.

• YouTube
• 55 minutes
• On-Demand
• Free Video

Benchmark and Critical Evaluation for ML Force Fields with Molecular Simulations

Explore benchmarking and critical evaluation of machine learning force fields in molecular simulations, focusing on realistic MD trajectories beyond force prediction accuracy.

• YouTube
• 1 hour 1 minute
• On-Demand
• Free Video

Score Matching via Differentiable Physics

Explore score matching via differentiable physics for inverse problems. Learn about combining physics operators with stochastic differential equations to generate plausible trajectories towards given end states.

• YouTube
• 1 hour 5 minutes
• On-Demand
• Free Video