google-research

TensorFlow code and pre-trained models for BERT

Google Research

TimesFM (Time Series Foundation Model) is a pretrained time-series foundation model developed by Google Research for time-series forecasting.

A playbook for systematically maximizing the performance of deep learning models.

arXiv LaTeX Cleaner: Easily clean the LaTeX code of your paper to submit to arXiv

Code for the paper "Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer"

SimCLRv2 - Big Self-Supervised Models are Strong Semi-Supervised Learners

Scenic: A Jax Library for Computer Vision Research and Beyond

A Code Release for Mip-NeRF 360, Ref-NeRF, and RawNeRF

Check out the new game server:

Official codebase used to develop Vision Transformer, SigLIP, MLP-Mixer, LiT and more.

ALBERT: A Lite BERT for Self-supervised Learning of Language Representations

FILM: Frame Interpolation for Large Motion, In ECCV 2022.

A data generation pipeline for creating semi-realistic synthetic multi-object videos with rich annotations such as instance segmentation masks, depth maps, and optical flow.

TabFM (Tabular Foundation Model) is a pretrained tabular foundation model developed by Google Research for tabular data regression and classification.

ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators

Unsupervised Data Augmentation (UDA)

Shared repository for open-sourced projects from the Google AI Language team.

Differentiable SDE solvers with GPU support and efficient sensitivity analysis.

Research language for array processing in the Haskell/ML family

Official repository for the "Big Transfer (BiT): General Visual Representation Learning" paper.

disentanglement_lib is an open-source library for research on learning disentangled representations.

Learning Latent Dynamics for Planning from Pixels

A simple method to perform semi-supervised learning with limited data.

End-to-end neural table-text understanding models.

[CVPR 2022 Oral] Official repository for "MAXIM: Multi-Axis MLP for Image Processing". SOTA for denoising, deblurring, deraining, dehazing, and enhancement.

Fast & Simple Resource-Constrained Learning of Deep Network Structure

DeepLab2 is a TensorFlow library for deep labeling, aiming to provide a unified and state-of-the-art TensorFlow codebase for dense pixel labeling tasks.

AndroidWorld is an environment and benchmark for autonomous agents

Official JAX implementation of MAGVIT: Masked Generative Video Transformer

AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty

Efficient Batched Reinforcement Learning in TensorFlow

Pix2Seq codebase: multi-tasks with generative modeling (autoregressive and diffusion)

[NeurIPS'21 Outstanding Paper] Library for reliable evaluation on RL and ML benchmarks, even with only a handful of seeds.

SEED RL: Scalable and Efficient Deep-RL with Accelerated Central Inference. Implements IMPALA and R2D2 algorithms in TF2 with SEED's architecture.

A dataset of datasets for learning to learn from few examples

BLEURT is a metric for Natural Language Generation based on transfer learning.

Long Range Arena for Benchmarking Efficient Transformers

A Configurable Recommender Systems Simulation Platform

A collection of Google research projects related to Federated Learning and Federated Analytics.

Code for Noisy Student Training. https://arxiv.org/abs/1911.04252

Dream to Control: Learning Behaviors by Latent Imagination

[CoRL '23] Dexterous piano playing with deep reinforcement learning.

A reimplementation of "The Lottery Ticket Hypothesis" (Frankle and Carbin) on MNIST.

NASBench: A Neural Architecture Search Dataset and Benchmark

Code release for CVPR'24 submission 'OmniGlue'

Original Implementation of Prompt Tuning from Lester, et al, 2021

XTREME is a benchmark for the evaluation of the cross-lingual generalization ability of pre-trained multilingual models that covers 40 typologically diverse languages and includes nine tasks.

A benchmark for the next generation of data-driven global weather models.

Transformers for Longer Sequences

Train robotic agents to learn pick and place with deep learning for vision-based manipulation in PyBullet. Transporter Nets, CoRL 2020.

Fast Differentiable Sorting and Ranking

Distributed machine learning infrastructure for large-scale robotics research

Offline Reinforcement Learning (aka Batch Reinforcement Learning) on Atari 2600 games

Multi-modal Content Creation Model Training Infrastructure including the FACT model (AI Choreographer) implementation.

Official Jax Implementation of MaskGIT

LEAF is a learnable alternative to audio features such as mel-filterbanks, that can be initialized as an approximation of mel-filterbanks, and then be trained for the task at hand, while using a...

Recipes for reproducing Analysis-Ready & Cloud Optimized (ARCO) ERA5 datasets.

[ECCV 2022] Official repository for "MaxViT: Multi-Axis Vision Transformer". SOTA foundation models for classification, detection, segmentation, image quality, and generative modeling...

Learning to Prompt (L2P) for Continual Learning @ CVPR22 and DualPrompt: Complementary Prompting for Rehearsal-free Continual Learning @ ECCV22

Compute receptive fields of your favorite convnets

PDFM Embeddings: location-based vectors for geo-spatial analysis.

Code for the paper "Defeating Prompt Injections by Design"

Semi-supervised learning for object detection

Shape, Pose, and Appearance from a Single Image via Bootstrapped Radiance Field Inversion

Understanding computation in artificial and biological recurrent networks through the lens of dynamical systems.

Official implementation of Implicit Behavioral Cloning, as described in our CoRL 2021 paper, see more at https://implicitbc.github.io/

Suite of human-collected datasets and a multi-task continuous control benchmark for open vocabulary visuolinguomotor learning.

Real-World RL Benchmark Suite

Splat-SLAM: Globally Optimized RGB-only SLAM with 3D Gaussians