A Scalable Approach for Partially Local Federated Learning
Training Machine Learning Models More Efficiently with Dataset Distillation
Interpretable Deep Learning for Time Series Forecasting
A Fast WordPiece Tokenization System
More Efficient In-Context Learning with GLaM
General and Scalable Parallelization for Neural Networks
Improving Vision Transformer Efficiency and Accuracy by Learning to Tokenize
Google at NeurIPS 2021
Evaluating Syntactic Abilities of Language Models
RLDS: An Ecosystem to Generate, Share, and Use Datasets in Reinforcement Learning
MURAL: Multimodal, Multi-task Retrieval Across Languages
Predicting Text Selections with Federated Learning
Decisiveness in Imitation Learning for Robots
Permutation-Invariant Neural Networks for Reinforcement Learning
Predicting Text Readability from Scrolling Interactions
RLiable: Towards Reliable Evaluation & Reporting in Reinforcement Learning
MetNet-2: Deep Learning for 12-Hour Precipitation Forecasting
An Open Source Vibrotactile Haptics Platform for On-Body Applications
Making Better Future Predictions by Watching Unlabeled Videos
Model Ensembles Are Faster Than You Think
Enhanced Sleep Sensing in Nest Hub
Google at EMNLP 2021
Improved On-Device ML on Pixel 6, with Neural Architecture Search
Another Step Towards Breakeven Fusion
Self-Supervised Reversibility-Aware Reinforcement Learning
GoEmotions: A Dataset for Fine-Grained Emotion Classification
Grammar Correction as You Type, on Pixel 6
Deciding Which Tasks Should Train Together in Multi-Task Neural Networks
Practical Differentially Private Clustering
Predicting Spreadsheet Formulas from Semi-structured Contexts
How Underspecification Presents Challenges for Machine Learning
SimVLM: Simple Visual Language Model Pre-training with Weak Supervision
Baselines for Uncertainty and Robustness in Deep Learning
An ML-Based Framework for COVID-19 Epidemiology
Self-Supervised Learning Advances Medical Image Classification
Google at ICCV 2021
Finding Complex Metal Oxides for Technology Advancement
Introducing FLAN: More generalizable Language Models with Instruction Fine-Tuning
FedJAX: Federated Learning Simulation with JAX
Efficient Partitioning of Road Networks
Improving Generalization in Reinforcement Learning using Policy Similarity Embeddings
High-Quality, Robust and Responsible Direct Speech-to-Speech Translation
Pathdreamer: A World Model for Indoor Navigation
Announcing WIT: A Wikipedia-Based Image-Text Dataset
Toward Fast and Accurate Neural Networks for Image Recognition
Revisiting Mask-Head Architectures for Novel Class Instance Segmentation
Music Conditioned 3D Dance Generation with AIST++
Personalized ASR Models from a Large and Diverse Disordered Speech Dataset
Discovering Anomalous Data with Self-Supervised Learning
Detecting Abnormal Chest X-rays using Deep Learning
Introducing Omnimattes: A New Approach to Matte Generation using Layered Neural Rendering
Recreating Natural Voices for People with Speech Impairments
SoundStream: An End-to-End Neural Audio Codec
Demonstrating the Fundamentals of Quantum Error Correction
The C4_200M Synthetic Dataset for Grammatical Error Correction
A Dataset Exploration Case Study with Know Your Data
Improved Detection of Elusive Polyps via Machine Learning
Two New Datasets for Conversational NLP: TimeDial and Disfl-QA
Google at ACL 2021
Mapping Africa’s Buildings with Satellite Imagery
Advances in TF-Ranking
Applying Advanced Speech Enhancement in Cochlear Implants
Multi-task Prediction of Organ Dysfunction in ICUs
Google at ICML 2021
High Fidelity Image Generation Using Diffusion Models
Speeding Up Reinforcement Learning with a New Physics Simulation Engine
From Vision to Language: Semi-supervised Learning in Action … at Scale
Reducing the Computational Cost of Deep Reinforcement Learning Research
Quickly Training Game-Playing Agents with Machine Learning
Take All Your Pictures to the Cleaners, with Google Photos Noise and Blur Reduction
Achieving Precision in Quantum Material Simulations
A Dataset for Studying Gender Bias in Translation
Improving Genomic Discovery with Machine Learning
Quantum Machine Learning and the Power of Data
Google at CVPR 2021
Learning an Accurate Physics Simulator via Adversarial Reinforcement Learning
A Step Toward More Inclusive People Annotations in the Open Images Extended Dataset
The Importance of A/B Testing in Robotics
FRILL: On-Device Speech Representations using TensorFlow-Lite
Using Variational Transformer Networks to Automate Document Layout Design
Extending Contrastive Learning to the Supervised Setting
Data Cascades in Machine Learning
Toward Generalized Sim-to-Real Transfer for Robot Learning
A Browsable Petascale Reconstruction of the Human Cortex
Cross-Modal Contrastive Learning for Text-to-Image Generation
Understanding Contextual Facial Expressions Across the Globe
KELM: Integrating Knowledge Graphs with Language Model Pre-training Corpora
Project Guideline: Enabling Those with Low Vision to Run Independently
Learning to Manipulate Deformable Objects
ALIGN: Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision
Accelerating Eye Movement Research for Wellness and Accessibility
Crisscrossed Captions: Semantic Similarity for Images and Text
Introducing FELIX: Flexible Text Editing Through Tagging and Insertion
Do Wide and Deep Networks Learn the Same Things?
Google at ICLR 2021
Flexible, Scalable, Differentiable Simulation of Recommender Systems with RecSim NG
Model-Based RL for Decentralized Multi-agent Navigation
Holistic Video Scene Understanding with ViP-DeepLab
