Discover google-deepmind/penzai Open Source project

Open source code for AlphaFold.

13,022

alphafold

TensorFlow-based neural network library

12,240

sonnet

StarCraft II Learning Environment

9,741

pysc2

Multi-Joint dynamics with Contact. A general purpose physics simulator.

7,963

mujoco

A customisable 3D platform for agent-based AI research

7,749

lab

7,080

graph_nets

Build Graph Nets in Tensorflow

graphcast

open_spiel

OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.

Learning to Learn in TensorFlow

4,159

learning-to-learn

alphageometry

dm_control

Google DeepMind's software stack for physics-based simulation and Reinforcement Learning environments, using MuJoCo.

A library of reinforcement learning components and agents

3,702

acme

TensorFlow Reinforcement Learning

3,463

trfl

JAX-based neural network library

3,139

dm-haiku

alphatensor

dnc

A TensorFlow implementation of the Differentiable Neural Computer.

Open weights LLM from Google DeepMind.

2,478

gemma

Monte Carlo tree search in JAX

2,275

mctx

code_contests

kinetics-i3d

Convolutional neural network model for video classification trained on the Kinetics dataset.

This dataset code generates mathematical question and answer pairs, from a range of question types at roughly school-level difficulty.

1,639

mathematics_dataset

Optax is a gradient processing and optimization library for JAX.

1,621

optax

bsuite is a collection of carefully-designed experiments that investigate core capabilities of a reinforcement learning (RL) agent

1,610

bsuite

educational

jraph

A Graph Neural Network Library in Jax

Question answering dataset featured in "Teaching Machines to Read and Comprehend

1,349

rc-data

1,285

tapnet

Tracking Any Point (TAP)

A collection of high-quality models for the MuJoCo physics engine, curated by Google DeepMind.

1,245

mujoco_menagerie

rlax

scalable_agent

A TensorFlow implementation of Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures.

RL research on Android devices.

980

android_env

This repository contains notebook implementations of the following Neural Process variants: Conditional Neural Processes (CNPs), Neural Processes (NPs), Attentive Neural Processes (ANPs).

977

neural-processes

Dramatron uses large language models to generate coherent scripts and screenplays.

969

dramatron

tree is a library for working with nested data structures

941

tree

Real-time behaviour synthesis with MuJoCo, using Predictive Control

925

mujoco_mpc

materials_discovery

xmanager

A platform for managing machine learning experiments

chex

reverb

Reverb is an efficient and easy-to-use data storage and transport system designed for machine learning research

alphadev

funsearch

pycolab

A highly-customisable gridworld game engine with some batteries included. Make your own gridworld games to test reinforcement learning agents!

An implementation of the Fermionic Neural Network for ab-initio electronic structure calculations

659

ferminet

hanabi_learning_environment is a research platform for Hanabi experiments.

651

hanabi-learning-environment

open_x_embodiment

recurrentgemma

Open weights language model from Google DeepMind, based on Griffin.

This is a suite of reinforcement learning environments illustrating various safety properties of intelligent agents.

584

ai-safety-gridworlds

A suite of test scenarios for multi-agent reinforcement learning.

577

meltingpot

Lua/Torch implementation of DQN (Nature, 2015)

576

dqn

Lua

546

ithaca

Restoring and attributing ancient texts using deep neural networks

distrax

surface-distance

Library to compute surface distance based performance metrics for segmentation tasks.

Benchmarking long-form factuality in large language models. Original code for our paper "Long-form factuality in large language models".

526

long-form-factuality

tracr

alphamissense

dsprites-dataset

Dataset to assess the disentanglement properties of unsupervised learning methods

A library for generative social simulation

476

concordia

This repository contains the NarrativeQA dataset. It includes the list of documents with Wikipedia summaries, links to full stories, and questions and answers.

472

narrativeqa

clrs

lab2d

A customisable 2D platform for agent-based AI research

DQN Zoo is a collection of reference implementations of reinforcement learning agents developed at DeepMind based on the Deep Q-Network (DQN) agent.

420

dqn_zoo

alphastar

dm_pix

PIX is an image processing library in JAX, for JAX.

Spriteworld: a flexible, configurable python-based reinforcement learning environment

376

spriteworld

367

mathematics_conjectures

A Python interface for reinforcement learning environments

362

dm_env

official code for "Large Language Models as Optimizers"

343

opro

Libraries, tools and tasks created and used at DeepMind Robotics.

340

dm_robotics

Dermatology ddx dataset, Jax implementations of Monte Carlo conformal prediction, plausibility regions and statistical annotation aggregation from our recent work on uncertain ground truth (TMLR'23 and ArXiv pre-print).

332

uncertain_ground_truth

We provide a pre-trained model for unconditional 19-step generation of CelebA-HQ images

332

spiral

TORAX: Tokamak transport simulation in JAX

327

torax

launchpad

leo

Implementation of Meta-Learning with Latent Embedding Optimization

A C++/Python implementation of the StreetLearn environment based on images from Street View, as well as a TensorFlow implementation of goal-driven navigation agents solving the task published in “Learning to Navigate in Cities Without a Map”, NeurIPS 2018

302

streetlearn

enn

gqn-datasets

Datasets used to train Generative Query Networks (GQNs) in the ‘Neural Scene Representation and Rendering’ paper.

Multi-object image datasets with ground-truth segmentation masks and generative factors.

269

multi_object_datasets

A algebraic word problem dataset, with multiple choice questions annotated with rationales.

251

AQuA

238

synjax

A code generation dataset for generating the code that implements Hearthstone and Magic The Gathering card effects.

238

card2code

236

grid-cells

Implementation of the supervised learning experiments in Vector-based navigation using grid-like representations in artificial agents, as published at https://www.nature.com/articles/s41586-018-0102-6

arnheim

torch-hdf5

Torch interface to HDF5 library

Lua

234

dm_memorytasks

A set of 13 diverse machine-learning tasks that require memory to solve.

Exploiting temporal context for 3D human pose estimation in the wild: 3D poses for the Kinetics dataset

221

Temporal-3D-Pose-Kinetics

Second Order Optimization and Curvature Estimation with K-FAC in JAX.

218

kfac-jax

DeepMind Alchemy task environment: a meta-reinforcement learning benchmark

209

dm_alchemy

neural_testbed

jmp

JMP is a Mixed Precision library for JAX.

perception_test

pg19

neural_networks_chomsky_hierarchy

Neural Networks and the Chomsky Hierarchy

Implementation of Spectral Inference Networks, ICLR 2019

174

xquad

174

spectral_inference_networks

Progressive matrices dataset, as described in: Measuring abstract reasoning in neural networks (Barrett*, Hill*, Santoro*, Morcos, Lillicrap), ICML2018

165

abstract-reasoning-matrices

162

xitari

This is the 0.4 release of the Arcade Learning Environment (ALE), a platform designed for AI research. ALE is based on Stella, an Atari 2600 VCS emulator.

Annotating tensor shapes using Python types

159

tensor_annotations

mc_gradients

symplectic-gradient-adjustment

A colab that implements the Symplectic Gradient Adjustment optimizer from "The mechanics of n-player differentiable games"