# Survival Instinct in Offline Reinforcement Learning

arXiv: https://arxiv.org/abs/2306.03286 · HTML: https://mlsanity.com/paper/2306.03286/

## Key facts

- Cited by 20 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 59 of its 72 references that are in the map.
- First posted on arXiv 5 Jun 2023.

## 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 59)

- [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)
- [D4RL: Datasets for Deep Data-Driven Reinforcement Learning](https://mlsanity.com/paper/2004.07219/) (2020)
- [Offline Reinforcement Learning with Implicit Q-Learning](https://mlsanity.com/paper/2110.06169/) (2021)
- [Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning](https://mlsanity.com/paper/1910.10897/) (2019)
- [Constrained Policy Optimization](https://mlsanity.com/paper/1705.10528/) (2017)
- [A Minimalist Approach to Offline Reinforcement Learning](https://mlsanity.com/paper/2106.06860/) (2021)
- [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)
- [COMBO: Conservative Offline Model-Based Policy Optimization](https://mlsanity.com/paper/2102.08363/) (2021)
- [Is Pessimism Provably Efficient for Offline RL?](https://mlsanity.com/paper/2012.15085/) (2020)
- [The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models](https://mlsanity.com/paper/2201.03544/) (2022)
- [Reward learning from human preferences and demonstrations in Atari](https://mlsanity.com/paper/1811.06521/) (2018)
- [Information-Theoretic Considerations in Batch Reinforcement Learning](https://mlsanity.com/paper/1905.00360/) (2019)
- [Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimism](https://mlsanity.com/paper/2103.12021/) (2021)
- [Bellman-consistent Pessimism for Offline Reinforcement Learning](https://mlsanity.com/paper/2106.06926/) (2021)
- [Inverse Reward Design](https://mlsanity.com/paper/1711.02827/) (2017)
- [Batch Policy Learning under Constraints](https://mlsanity.com/paper/1903.08738/) (2019)
- [Offline-to-Online Reinforcement Learning via Balanced Replay and Pessimistic Q-Ensemble](https://mlsanity.com/paper/2107.00591/) (2021)
- [Deep Reinforcement Learning and the Deadly Triad](https://mlsanity.com/paper/1812.02648/) (2018)
- [Pessimistic Model-based Offline Reinforcement Learning under Partial Coverage](https://mlsanity.com/paper/2107.06226/) (2021)
- [Offline Reinforcement Learning with Realizability and Single-policy Concentrability](https://mlsanity.com/paper/2202.04634/) (2022)
- [Safe Reinforcement Learning in Constrained Markov Decision Processes](https://mlsanity.com/paper/2008.06626/) (2020)
- [RAMBO-RL: Robust Adversarial Model-Based Offline Reinforcement Learning](https://mlsanity.com/paper/2204.12581/) (2022)
- [Adversarially Trained Actor Critic for Offline Reinforcement Learning](https://mlsanity.com/paper/2202.02446/) (2022)
- [AI Safety Gridworlds](https://mlsanity.com/paper/1711.09883/) (2017)
- [Don't Change the Algorithm, Change the Data: Exploratory Data for Offline Reinforcement Learning](https://mlsanity.com/paper/2201.13425/) (2022)
- [Robust Reinforcement Learning using Offline Data](https://mlsanity.com/paper/2208.05129/) (2022)
- [Constraints Penalized Q-learning for Safe Offline Reinforcement Learning](https://mlsanity.com/paper/2107.09003/) (2021)

## Built on by (most cited first, 20 of 20)

- [Towards Robust Offline Reinforcement Learning under Diverse Data Corruption](https://mlsanity.com/paper/2310.12955/) (2023)
- [Listwise Reward Estimation for Offline Preference-based Reinforcement Learning](https://mlsanity.com/paper/2408.04190/) (2024)
- [Unsupervised Behavior Extraction via Random Intent Priors](https://mlsanity.com/paper/2310.18687/) (2023)
- [OASIS: Conditional Distribution Shaping for Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2407.14653/) (2024)
- [Uncertainty-based Offline Variational Bayesian Reinforcement Learning for Robustness under Diverse Data Corruptions](https://mlsanity.com/paper/2411.00465/) (2024)
- [Tackling Data Corruption in Offline Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2407.04285/) (2024)
- [Offline Reinforcement Learning for Learning to Dispatch for Job Shop Scheduling](https://mlsanity.com/paper/2409.10589/) (2024)
- [Dual-Robust Cross-Domain Offline Reinforcement Learning Against Dynamics Shifts](https://mlsanity.com/paper/2512.02486/) (2025)
- [Universal Black-Box Reward Poisoning Attack against Offline Reinforcement Learning](https://mlsanity.com/paper/2402.09695/) (2024)
- [Adversarially Robust Decision Transformer](https://mlsanity.com/paper/2407.18414/) (2024)
- [SaVeR: Optimal Data Collection Strategy for Safe Policy Evaluation in Tabular MDP](https://mlsanity.com/paper/2406.02165/) (2024)
- [Hindsight Preference Learning for Offline Preference-based Reinforcement Learning](https://mlsanity.com/paper/2407.04451/) (2024)
- [CLARIFY: Contrastive Preference Reinforcement Learning for Untangling Ambiguous Queries](https://mlsanity.com/paper/2506.00388/) (2025)
- [Adversarial Policy Optimization for Offline Preference-based Reinforcement Learning](https://mlsanity.com/paper/2503.05306/) (2025)
- [ADG: Ambient Diffusion-Guided Dataset Recovery for Corruption-Robust Offline Reinforcement Learning](https://mlsanity.com/paper/2505.23871/) (2025)
- [Binary Reward Labeling: Bridging Offline Preference and Reward-Based Reinforcement Learning](https://mlsanity.com/paper/2406.10445/) (2024)
- [From Reward-Free Representations to Preferences: Rethinking Offline Preference-Based Reinforcement Learning](https://mlsanity.com/paper/2606.01123/) (2026)
- [TROFI: Trajectory-Ranked Offline Inverse Reinforcement Learning](https://mlsanity.com/paper/2506.22008/) (2025)
- [Generalizing Beyond Suboptimality: Offline Reinforcement Learning Learns Effective Scheduling through Random Solutions](https://mlsanity.com/paper/2509.10303/) (2025)
- [OPRIDE: Offline Preference-based Reinforcement Learning via In-Dataset Exploration](https://mlsanity.com/paper/2604.02349/) (2026)

## Abstract (arXiv)

We present a novel observation about the behavior of offline reinforcement learning (RL) algorithms: on many benchmark datasets, offline RL can produce well-performing and safe policies even when trained with "wrong" reward labels, such as those that are zero everywhere or are negatives of the true rewards. This phenomenon cannot be easily explained by offline RL's return maximization objective. Moreover, it gives offline RL a degree of robustness that is uncharacteristic of its online RL counterparts, which are known to be sensitive to reward design. We demonstrate that this surprising robustness property is attributable to an interplay between the notion of pessimism in offline RL algorithms and certain implicit biases in common data collection practices. As we prove in this work, pessimism endows the agent with a "survival instinct", i.e., an incentive to stay within the data support in the long term, while the limited and biased data coverage further constrains the set of survival policies. Formally, given a reward class -- which may not even contain the true reward -- we identify conditions on the training data distribution that enable offline RL to learn a near-optimal and safe policy from any reward within the class. We argue that the survival instinct should be taken into account when interpreting results from existing offline RL benchmarks and when creating future ones. Our empirical and theoretical results suggest a new paradigm for RL, whereby an agent is nudged to learn a desirable behavior with imperfect reward but purposely biased data coverage.