Jascha Sohl-Dickstein

Diffusion-Probabilistic-Models. Reference implementation for Deep Unsupervised Learning using Nonequilibrium Thermodynamics

fractal. The boundary of neural network trainability is fractal

Sum-of-Functions-Optimizer. Implements SFO minibatch optimizer in Python and MATLAB, and reproduces figures from paper.

Minimum-Probability-Flow-Learning. Matlab code implementing Minimum Probability Flow Learning.

LAHMC. Look Ahead Hamiltonian Monte Carlo

Hamiltonian-Annealed-Importance-Sampling. Matlab code implementing Hamiltonian Annealed Importance Sampling for importance weight, partition function, and log likelihood estimation for models with continuous state spaces

DrBrainlove. Code for regenerating and manipulating label stickers for Dr. Brainlove art car and science outreach project.

NL-Augmenter. NL-Augmenter 🦎 → 🐍 A Collaborative Repository of Natural Language Transformations

guided-evolutionary-strategies. Guided Evolutionary Strategies

jax. Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

learned_optimization. Python