Pith. sign in

REVIEW 26 cited by

Conservative Q-Learning for Offline Reinforcement Learning

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2006.04779 v3 pith:BS4VGKKU submitted 2020-06-08 cs.LG stat.ML

Conservative Q-Learning for Offline Reinforcement Learning

classification cs.LG stat.ML
keywords learningofflinepolicyconservativeq-learningvaluechallengecomplex
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Effectively leveraging large, previously collected datasets in reinforcement learning (RL) is a key challenge for large-scale real-world applications. Offline RL algorithms promise to learn effective policies from previously-collected, static datasets without further interaction. However, in practice, offline RL presents a major challenge, and standard off-policy RL methods can fail due to overestimation of values induced by the distributional shift between the dataset and the learned policy, especially when training on complex and multi-modal data distributions. In this paper, we propose conservative Q-learning (CQL), which aims to address these limitations by learning a conservative Q-function such that the expected value of a policy under this Q-function lower-bounds its true value. We theoretically show that CQL produces a lower bound on the value of the current policy and that it can be incorporated into a policy learning procedure with theoretical improvement guarantees. In practice, CQL augments the standard Bellman error objective with a simple Q-value regularizer which is straightforward to implement on top of existing deep Q-learning and actor-critic implementations. On both discrete and continuous control domains, we show that CQL substantially outperforms existing offline RL methods, often learning policies that attain 2-5 times higher final return, especially when learning from complex and multi-modal data distributions.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 26 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Offline Reinforcement Learning with Implicit Q-Learning

    cs.LG 2021-10 unverdicted novelty 8.0

    IQL achieves policy improvement in offline RL by implicitly estimating optimal action values through state-conditional upper expectiles of value functions, without querying Q-functions on out-of-distribution actions.

  2. Decoupling KL and Trajectories: A Unified Perspective for SFT, DAgger, Offline RL, and OPD in LLM Distillation

    cs.LG 2026-05 unverdicted novelty 7.0

    Decoupling prefix source from token-level KL direction in autoregressive sequence KL yields four objectives unifying SFT, DAgger, offline RL and OPD, with KL mixing and entropy-gated curriculum improving math reasonin...

  3. Weights or Skills? A Survey of Robot-Learning Techniques: from Action-Predicting Weights to Robots that Write their Own Skills

    cs.RO 2026-08 conditional novelty 6.0

    A taxonomy of robot learning on a weights-versus-skills axis, with a five-rung self-improvement ladder whose top cell (feedback plus memory plus search) holds only a few recent systems.

  4. Good Rankers, Bad Objectives: Bilinear Contrastive Critics under Expressive Policy Search

    cs.LG 2026-07 conditional novelty 6.0

    Bilinear contrastive critics remain good compatibility rankers but are unsafe to maximize for action selection; cosine bounding does not fix value decalibration, while Bellman TD-Q does.

  5. Support-Constrained RL Enables Real-World Policy Improvement without Real-World Experience

    cs.RO 2026-06 unverdicted novelty 6.0

    SCORE constrains sim RL to the support of a real-data policy via flow steering, raising average success on eight dexterous tasks from 37.8% to 89.9%.

  6. Reward-free Pretraining for Reinforcement Learning via Occupancy Coverage Maximization

    cs.LG 2026-06 unverdicted novelty 6.0

    ROVER pretrains transferable exploration policies by maximizing occupancy coverage with a learned resolvent world model and virtual sink state, outperforming baselines on sparse navigation tasks.

  7. RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking

    cs.AI 2026-05 unverdicted novelty 6.0

    RankQ adds a self-supervised ranking loss to Q-learning to learn structured action orderings, yielding competitive or better performance than prior methods on D4RL benchmarks and large gains in vision-based robot fine-tuning.

  8. An adaptive variance estimator for relative sparsity

    stat.ME 2026-05 unverdicted novelty 6.0

    A new adaptive variance estimator for relative sparsity coefficients is introduced that fully utilizes the prior asymptotic normality theorem and incorporates variable selection effects.

  9. Feedback-Normalized Developer Memory for Reinforcement-Learning Coding Agents: A Safety-Gated MCP Architecture

    cs.SE 2026-05 unverdicted novelty 6.0

    RL Developer Memory is a feedback-normalized, safety-gated memory architecture for RL coding agents that logs contextual decisions and applies conservative off-policy gates to maintain 80% decision accuracy and full h...

  10. JD-BP: A Joint-Decision Generative Framework for Auto-Bidding and Pricing

    cs.GT 2026-04 unverdicted novelty 6.0

    JD-BP jointly generates bids and pricing corrections via generative models, memory-less return-to-go, trajectory augmentation, and energy-based DPO to improve auto-bidding performance despite prediction errors and latency.

  11. JD-BP: A Joint-Decision Generative Framework for Auto-Bidding and Pricing

    cs.GT 2026-04 reject novelty 6.0

    A generative auto-bidding model that jointly emits a bid and an additive price correction outperforms bid-only baselines on AuctionNet and online at JD.com.

  12. Simulation Distillation: Pretraining World Models in Simulation for Rapid Real-World Adaptation

    cs.RO 2026-03 unverdicted novelty 6.0

    SimDist pretrains world models in simulation and adapts them to real-world robots by updating only the latent dynamics model, enabling rapid improvement on contact-rich tasks where prior methods fail.

  13. The hidden risks of temporal resampling in clinical reinforcement learning

    cs.LG 2026-02 conditional novelty 6.0

    Resampling clinical time series into uniform bins for offline RL reduces performance by up to 60% and causes retrospective evaluations to overestimate returns by 1.5-3x versus unprocessed data.

  14. DAWM: Diffusion Action World Models for Offline Reinforcement Learning via Action-Inferred Transitions

    cs.LG 2025-09 unverdicted novelty 6.0

    DAWM introduces a modular diffusion world model with an inverse dynamics model to produce complete synthetic transitions that improve conservative offline RL algorithms like TD3BC and IQL on D4RL tasks.

  15. Generative Sequential Notification Optimization via Multi-Objective Decision Transformers

    cs.LG 2025-09 conditional novelty 6.0

    A Decision Transformer with quantile-regression return prompts improved notification decisions at LinkedIn, boosting sessions by 0.72% over the deployed CQL baseline in a live A/B test.

  16. What Matters in Learning from Offline Human Demonstrations for Robot Manipulation

    cs.RO 2021-08 accept novelty 6.0

    A comprehensive benchmark study of offline imitation learning methods on multi-stage robot manipulation tasks identifies key sensitivities to algorithm design, data quality, and stopping criteria while releasing all d...

  17. Guided Action Flow: Q-Guided Inference for Flow-Matching Vision-Language-Action Policies

    cs.RO 2026-07 conditional novelty 5.5

    Q-guided reverse-time flow sampling with a rollout-trained action-chunk critic improves frozen SmolVLA success on LIBERO, with strong single-task gains but only modest locked held-out transfer.

  18. Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models

    cs.LG 2026-06 unverdicted novelty 5.0

    Higher conservatism in offline DPO training of Qwen3-14B monotonically increases reward-hacking damage (Goodhart gap AUGC) during online adaptation on GSM8K.

  19. From Bootstrapping to Sequence Modeling: A Unified Generative Framework for Personalized Landing-Page Modeling

    cs.IR 2026-06 unverdicted novelty 5.0

    GLAN replaces CQL bootstrapping with Decision Transformer sequence modeling for PLPM, using global inter-day (L-RTG) and local session (HRM) modules to achieve +0.158% DAU and +0.108% LT gains in Kuaishou online tests.

  20. Abstraction for Offline Goal-Conditioned Reinforcement Learning

    cs.LG 2026-05 unverdicted novelty 5.0

    Introduces relativised options and hierarchical abstraction to reuse experience across similar contexts in offline GCRL, with two algorithms demonstrating performance gains.

  21. ISEP: Implicit Support Expansion for Offline Reinforcement Learning via Stochastic Policy Optimization

    cs.LG 2026-05 unverdicted novelty 5.0

    ISEP expands action support in offline RL via value interpolation between data and policy samples, then uses stochastic policy optimization to avoid mode collapse in the resulting multimodal objective.

  22. RankQ: Offline-to-Online Reinforcement Learning via Self-Supervised Action Ranking

    cs.AI 2026-05 unverdicted novelty 5.0

    RankQ augments temporal-difference Q-learning with a multi-term self-supervised ranking loss to enforce structured action ordering, yielding competitive or better results than prior methods on D4RL and large gains in ...

  23. DVPO: Distributional Value Modeling-based Policy Optimization for LLM Post-Training

    cs.LG 2025-12 unverdicted novelty 5.0

    DVPO learns token-level value distributions and uses asymmetric risk regularization to contract lower tails while expanding upper tails, outperforming PPO and GRPO under noisy supervision in dialogue, math, and QA tasks.

  24. Value Flows

    cs.LG 2025-10 reject novelty 5.0

    Value Flows fits the full return distribution in RL with a flow-matching critic and reweights its learning objective by estimated return variance; the central theoretical guarantee does not follow from the stated equations.

  25. Guided Action Flow: Q-Guided Inference for Flow-Matching Vision-Language-Action Policies

    cs.RO 2026-07 unverdicted novelty 4.0

    Guided Action Flow applies a rollout-trained critic to steer frozen flow-matching VLA policies at inference time via action gradients, reporting success rate gains on LIBERO manipulation tasks.

  26. Reinforcement Learning: From Algorithms To Foundation Models

    cs.AI 2026-07 conditional novelty 3.0

    A dissertation uniting the author's published results: non-exploitable Nash-DQN policies and the FightLadder benchmark for games, plus diffusion/consistency-model world models for RL — a compilation rather than new results.