[NeurIPS 2021] SNIPS: Solving Noisy Inverse Problems Stochastically

The code enables to perform Bayesian inference in an efficient manner through the use of Hamiltonian Neural Networks (HNNs), Deep Neural Networks (DNNs), Neural ODEs, and Symplectic Neural Networks (SympNets) used with state-of-the-art sampling schemes like Hamiltonian Monte Carlo (HMC) and the No-U-Turn-Sampler (NUTS).

A demo shows how to combine Langevin dynamics with score matching for generative models.

Sampling-based approach to analyse neural networks using TensorFlow

Python solver for the Brownian, Stochastic, or Noisy Differential Equations

A python code to calculate the Brownian motion of colloidal particles in a time varying force field.

Noise-conditional score networks for music composition by annealed Langevin dynamics

Utilities for determining maximum tolerable timesteps. See https://doi.org/10.3390/e20050318

Finding Optimal Langevin Inferred Equations

The official code release for "More Efficient Randomized Exploration for Reinforcement Learning via Approximate Sampling", Reinforcement Learning Conference (RLC) 2024

Variationally enhanced sampling for single-particle langevin dynamics with neural network bias potentials and path collective variables. Based on OpenMM + PyTorch.

Python module for the evaluation of probability densities and exit rates in a tubular ensemble

The GitHub repository for "Accelerating Approximate Thompson Sampling with Underdamped Langevin Monte Carlo", AISTATS 2024.

Bayesian Neural Network (BNN) implementations based on Langevin Dynamics and tested on real-world data

A simulation framework for nonequilibrium statistical physics

Langevin dynamics simulation of bacteriophage-bacteria interaction

The GitHub repository for "Constrained Exploration via Reflected Replica Exchange Stochastic Gradient Langevin Dynamics", ICML 2024

Python module for (symbolic) evaluation of the short-time Fokker-Planck propagator to arbitrary accuracy