Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

ScienceWorld is a text-based virtual environment centered around accomplishing tasks from the standardized elementary science curriculum.

A latent text-to-image diffusion model

PyTorch implementation for "Training and Inference on Any-Order Autoregressive Models the Right Way", NeurIPS 2022 Oral, TPM 2023 Best Paper Honorable Mention

(NeurIPS '21 Spotlight) IQ-Learn: Inverse Q-Learning for Imitation

Implementations of IQL, QMIX, VDN, COMA, QTRAN, MAVEN, CommNet, DyMA-CL, and G2ANet on SMAC, the decentralised micromanagement scenario of StarCraft II

Gridworld for MARL experiments

Official implementation of "On GANs and GMMs"

Code for Neural Architecture Search without Training (ICML 2021)

Sliced Wasserstein Distance for Learning Gaussian Mixture Models

The Additive Margin SincNet (AM-SincNet) is a new approach for speaker recognition problems which is based in the neural network architecture SincNet and the additive margin softmax (AM-Softmax) lo…

Code for confidence sequences for sampling without replacement (NeurIPS, 2020). https://arxiv.org/pdf/2006.04347.pdf

A PyTorch implementation of "Multimodal Generative Models for Scalable Weakly-Supervised Learning" (https://arxiv.org/abs/1802.05335)

Multimodal Mixture-of-Experts VAE

Efficient Graph Generation with Graph Recurrent Attention Networks, Deep Generative Model of Graphs, Graph Neural Networks, NeurIPS 2019

GraphGen: A Scalable Approach to Domain-agnostic Labeled Graph Generation

PyTorch Re-Implementation of "Generating Sentences from a Continuous Space" by Bowman et al 2015 https://arxiv.org/abs/1511.06349