# Offline Q-Learning on Diverse Multi-Task Data Both Scales And Generalizes

arXiv: https://arxiv.org/abs/2211.15144 · HTML: https://mlsanity.com/paper/2211.15144/

## Key facts

- Cited by 73 papers in the map (data through 14 Oct 2026).
- Placed in Offline RL algorithms, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 54 of its 75 references that are in the map.
- First posted on arXiv 28 Nov 2022.

## Placed in

- Offline RL and imitation learning from data / [Offline RL algorithms](https://mlsanity.com/offlinerl/offline-rl-algorithms/)

## Builds on (most cited first, 30 of 54)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Conservative Q-Learning for Offline Reinforcement Learning](https://mlsanity.com/paper/2006.04779/) (2020)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Group Normalization](https://mlsanity.com/paper/1803.08494/) (2018)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [Off-Policy Deep Reinforcement Learning without Exploration](https://mlsanity.com/paper/1812.02900/) (2018)
- [A Minimalist Approach to Offline Reinforcement Learning](https://mlsanity.com/paper/2106.06860/) (2021)
- [IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures](https://mlsanity.com/paper/1802.01561/) (2018)
- [Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction](https://mlsanity.com/paper/1906.00949/) (2019)
- [A Generalist Agent](https://mlsanity.com/paper/2205.06175/) (2022)
- [Deep Reinforcement Learning at the Edge of the Statistical Precipice](https://mlsanity.com/paper/2108.13264/) (2021)
- [Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning](https://mlsanity.com/paper/1910.00177/) (2019)
- [A Distributional Perspective on Reinforcement Learning](https://mlsanity.com/paper/1707.06887/) (2017)
- [AWAC: Accelerating Online Reinforcement Learning with Offline Datasets](https://mlsanity.com/paper/2006.09359/) (2020)
- [Behavior Regularized Offline Reinforcement Learning](https://mlsanity.com/paper/1911.11361/) (2019)
- [MOPO: Model-based Offline Policy Optimization](https://mlsanity.com/paper/2005.13239/) (2020)
- [MOReL : Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2005.05951/) (2020)
- [Policy Distillation](https://mlsanity.com/paper/1511.06295/) (2015)
- [COMBO: Conservative Offline Model-Based Policy Optimization](https://mlsanity.com/paper/2102.08363/) (2021)
- [Revisiting the Arcade Learning Environment: Evaluation Protocols and Open Problems for General Agents](https://mlsanity.com/paper/1709.06009/) (2017)
- [Masked Visual Pre-training for Motor Control](https://mlsanity.com/paper/2203.06173/) (2022)

## Built on by (most cited first, 30 of 73)

- [Learning Universal Policies via Text-Guided Video Generation](https://mlsanity.com/paper/2302.00111/) (2023)
- [TD-MPC2: Scalable, Robust World Models for Continuous Control](https://mlsanity.com/paper/2310.16828/) (2023)
- [Foundation Models for Decision Making: Problems, Methods, and Opportunities](https://mlsanity.com/paper/2303.04129/) (2023)
- [Bigger, Better, Faster: Human-level Atari with human-level efficiency](https://mlsanity.com/paper/2305.19452/) (2023)
- [Stop Regressing: Training Value Functions via Classification for Scalable Deep RL](https://mlsanity.com/paper/2403.03950/) (2024)
- [Revisiting the Minimalist Approach to Offline Reinforcement Learning](https://mlsanity.com/paper/2305.09836/) (2023)
- [Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learning](https://mlsanity.com/paper/2305.18459/) (2023)
- [Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control](https://mlsanity.com/paper/2405.16158/) (2024)
- [STEVE-1: A Generative Model for Text-to-Behavior in Minecraft](https://mlsanity.com/paper/2306.00937/) (2023)
- [Mixtures of Experts Unlock Parameter Scaling for Deep RL](https://mlsanity.com/paper/2402.08609/) (2024)
- [Horizon Reduction Makes RL Scalable](https://mlsanity.com/paper/2506.04168/) (2025)
- [Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models](https://mlsanity.com/paper/2504.11054/) (2025)
- [Anti-Exploration by Random Network Distillation](https://mlsanity.com/paper/2301.13616/) (2023)
- [1000 Layer Networks for Self-Supervised RL: Scaling Depth Can Enable New Goal-Reaching Capabilities](https://mlsanity.com/paper/2503.14858/) (2025)
- [In value-based deep reinforcement learning, a pruned network is a good network](https://mlsanity.com/paper/2402.12479/) (2024)
- [True Knowledge Comes from Practice: Aligning LLMs with Embodied Environments via Reinforcement Learning](https://mlsanity.com/paper/2401.14151/) (2024)
- [Offline Actor-Critic Reinforcement Learning Scales to Large Models](https://mlsanity.com/paper/2402.05546/) (2024)
- [The Generalization Gap in Offline Reinforcement Learning](https://mlsanity.com/paper/2312.05742/) (2023)
- [Reinformer: Max-Return Sequence Modeling for Offline RL](https://mlsanity.com/paper/2405.08740/) (2024)
- [Stabilizing Contrastive RL: Techniques for Robotic Goal Reaching from Offline Data](https://mlsanity.com/paper/2306.03346/) (2023)
- [Bigger, Regularized, Categorical: High-Capacity Value Functions are Efficient Multi-Task Learners](https://mlsanity.com/paper/2505.23150/) (2025)
- [Stable Gradients for Stable Learning at Scale in Deep Reinforcement Learning](https://mlsanity.com/paper/2506.15544/) (2025)
- [floq: Training Critics via Flow-Matching for Scaling Compute in Value-Based RL](https://mlsanity.com/paper/2509.06863/) (2025)
- [Zero-Shot Reinforcement Learning from Low Quality Data](https://mlsanity.com/paper/2309.15178/) (2023)
- [Understanding, Predicting and Better Resolving Q-Value Divergence in Offline-RL](https://mlsanity.com/paper/2310.04411/) (2023)
- [Digi-Q: Learning Q-Value Functions for Training Device-Control Agents](https://mlsanity.com/paper/2502.15760/) (2025)
- [Searching for High-Value Molecules Using Reinforcement Learning and Transformers](https://mlsanity.com/paper/2310.02902/) (2023)
- [Merging Decision Transformers: Weight Averaging for Forming Multi-Task Policies](https://mlsanity.com/paper/2303.07551/) (2023)
- [Contrastive Imitation Learning for Language-guided Multi-Task Robotic Manipulation](https://mlsanity.com/paper/2406.09738/) (2024)
- [AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers](https://mlsanity.com/paper/2411.11188/) (2024)

## Abstract (arXiv)

The potential of offline reinforcement learning (RL) is that high-capacity models trained on large, heterogeneous datasets can lead to agents that generalize broadly, analogously to similar advances in vision and NLP. However, recent works argue that offline RL methods encounter unique challenges to scaling up model capacity. Drawing on the learnings from these works, we re-examine previous design choices and find that with appropriate choices: ResNets, cross-entropy based distributional backups, and feature normalization, offline Q-learning algorithms exhibit strong performance that scales with model capacity. Using multi-task Atari as a testbed for scaling and generalization, we train a single policy on 40 games with near-human performance using up-to 80 million parameter networks, finding that model performance scales favorably with capacity. In contrast to prior work, we extrapolate beyond dataset performance even when trained entirely on a large (400M transitions) but highly suboptimal dataset (51% human-level performance). Compared to return-conditioned supervised approaches, offline Q-learning scales similarly with model capacity and has better performance, especially when the dataset is suboptimal. Finally, we show that offline Q-learning with a diverse dataset is sufficient to learn powerful representations that facilitate rapid transfer to novel games and fast online learning on new variations of a training game, improving over existing state-of-the-art representation learning approaches.