This paper bridges offline and online RL by studying policy finetuning with a reference policy.
problem Sample-efficient reinforcement learning in online and offline settings.
method Design of policy finetuning algorithms and analysis of sample complexity.
result Theoretical analysis shows that the optimal policy finetuning algorithm is either offline reduction or purely online RL.
New method selects best offline RL policies from logged data.
problem Hyperparameter selection challenges offline RL.
method Offline hyperparameter selection for RL algorithms.
result Reliable ranking and selection of policies across hyperparameters.
Hybrid RL algorithm combines offline and online data for robust and efficient policy learning.
problem Combining robust on-policy methods with efficient offline data for hybrid RL.
method Integrates off-policy training on offline data into on-policy NPG framework.
result Achieves state-of-the-art theoretical guarantees and maintains on-policy NPG guarantees.
MOReL learns offline RL policies using pessimistic MDPs.
problem Offline RL's data efficiency and velocity.
method Two-step process: learn P-MDP and near-optimal policy in it.
result MOReL is minimax optimal and matches state-of-the-art results.
A new method combines online and offline learning to tackle contextual bandits with missing action support.
problem Learning optimal policies with logged data when the logging policy has deficient support.
method Hybrid approach using online exploration to exploit supported actions and offline learning to avoid unnecessary explorations.
result Determines an optimal policy with theoretical guarantees using minimal online explorations.
BCPO optimizes offline RL policies by converting uncertainty into conservative bounds.
problem Offline RL's fragility under distribution shifts and model errors.
method Bayesian approach with credible lower bounds and KL regularization.
result BCPO yields an uncertainty-calibrated policy that avoids exploiting model errors.
A3RL combines online and offline RL with active sampling to improve policy learning.
problem Combining online and offline RL for sample efficiency and robustness.
method A3RL uses a confidence-aware Active Advantage Aligned (A3) sampling strategy to prioritize data from both online and offline sources.
result A3RL outperforms competing online RL techniques that use offline data.
Model-Based Offline Planning (MBOP) learns models from offline data to control systems directly.
problem Training RL policies from offline data without direct system interaction.
method Generates models from offline data and uses planning to control the system.
result Near-optimal policies found for simulated systems with minimal real-time interaction.
Paper addresses offline policy evaluation in RL, achieving near-optimal bounds for various policy classes.
problem Evaluate all policies in a class simultaneously for offline RL.
method Uniform convergence in OPE for various policy classes, achieving optimal episode complexity.
result Achieves optimal episode complexity of O(H^3/d_mε^2) for identifying ε-optimal policies.
ESRL uses uncertainty quantification to learn safe, optimal policies in offline RL.
problem Challenges in interpreting and measuring uncertainty of learned policies in offline RL.
method Expert-Supervised Reinforcement Learning (ESRL) framework that uses hypothesis testing and posterior distributions.
result The framework can learn safe and optimal policies with theoretical guarantees and independent sample efficiency.
CODA resolves coordination issues in offline multi-agent reinforcement learning.
problem Coordination failure in offline multi-agent reinforcement learning.
method Diffusion-based multi-agent trajectory generator for data augmentation.
result CODA resolves coordination pathologies in continuous polynomial games and complex benchmarks.
Algorithm improves reinforcement learning policies using offline data.
problem Improving reinforcement learning policies with limited online data.
method Designs a single non-reactive policy using offline data with provable guarantees.
result Algorithm achieves better policy quality with less online data.
New offline RL study shows exponential sample requirement for accurate policy evaluation.
problem Understanding statistical limits of offline RL with linear function approximation.
method Analyzes necessary representational and distributional conditions for sample-efficient offline reinforcement learning.
result Even with realizability and good feature coverage, offline RL requires exponential samples for accurate policy evaluation.
This work characterizes conditions for offline policy evaluation in reinforcement learning.
problem Understanding when classical methods succeed in offline policy evaluation for linear function approximation.
method Control-theoretic and linear-algebraic conditions for classical methods (FQI and LSTD).
result A precise hierarchy of regimes under which these estimators succeed, and a complete picture of their behavior.
Proposes a method to learn policies from offline data with reduced bias.
problem Learning policies from offline data with reduced bias and complexity constraints.
method Cross-fitted debiasing device for policy learning from offline data.
result Achieves N \sqrt N N regret for complex policy classes with a product-of-errors nuisance remainder. CQL (ReDS) learns from varied driving behaviors, improving offline RL performance.
problem Learning from datasets with non-uniform variability in behavior policies.
method Reweighting the data distribution to allow per-state flexibility in following the behavior policy.
result CQL (ReDS) improves performance in various offline RL tasks.
MOPO optimizes offline RL by penalizing dynamics uncertainty.
problem Learning policies from offline data with distributional shift.
method Modify model-based RL to avoid distributional shift.
result MOPO outperforms model-free and standard model-based RL.
New method tackles safe reinforcement learning from offline data.
problem Learn optimal policies from fixed data while adhering to safety constraints.
method Combines offline RL with online optimization to minimize cumulative cost.
result Proves approximate optimality of the approach under certain conditions.
New methods for evaluating and optimizing policies in offline RL with unobserved confounders.
problem Evaluating and optimizing policies in the presence of unobserved confounders.
method Characterized settings and algorithms for consistent value estimates and lower bounds, with sample complexity guarantees.
result Proved local convergence guarantees for offline policy improvement.
Paper tackles offline SSP with value iteration for policy evaluation and learning.
problem Goal-oriented RL with offline data and cost minimization.
method Simple value iteration algorithms for OPE and offline policy learning.
result Strong instance-dependent bounds implying near-minimax optimal worst-case bounds.
A new policy switching technique improves offline RL performance.
problem Challenges in adapting off-policy algorithms to different datasets and tasks.
method Combines off-policy RL and BC, using epistemic uncertainty for policy switching.
result Outperforms individual algorithms and state-of-the-art methods on benchmarks.
Paper tackles efficient evaluation of natural stochastic policies in offline RL.
problem Efficiency issues in evaluating natural stochastic policies due to unknown evaluation policy.
method Derive efficiency bounds for tilting and modified treatment policies, propose nonparametric estimators.
result Proposed estimators attain efficiency bounds under lax conditions and enjoy partial double robustness.
CQL learns conservative Q-functions to improve offline RL performance.
problem Leveraging large, static datasets in reinforcement learning without further interaction.
method Conservative Q-learning (CQL) which learns a conservative Q-function to lower-bound policy values.
result CQL substantially outperforms existing offline RL methods, often achieving 2-5 times higher final returns.
Offline RL policies should adapt to unknown aspects of the environment.
problem Uncertainty in offline RL datasets leads to suboptimal policies.
method Adaptive policies that consider all transitions seen so far, solving an implicit POMDP.
result Optimal adaptive policies improve offline RL performance.
CPPO learns policies from partial offline data in MDPs with structural assumptions.
problem Offline Reinforcement Learning with partial coverage assumption.
method Constrained Pessimistic Policy Optimization (CPPO) using a function class and model class constraint.
result CPPO achieves PAC guarantee with partial coverage, learning competitive policies.
This work bridges offline RL and DRL to address distributional shift.
problem Distributional shift in offline RL due to difference in state-action visitation distributions.
method Proposes offline RL algorithms using DRL framework, characterizes sample complexity under single policy concentrability.
result Demonstrates superior performance of proposed algorithms through simulations.
Critic-regularized regression improves offline RL performance.
problem Poor performance of off-policy algorithms in offline RL.
method Critic-regularized regression (CRR) for policy learning from fixed datasets.
result CRR outperforms state-of-the-art offline RL algorithms significantly.
Develops a method to efficiently use offline data for RL policy optimization.
problem Lack of online data for offline RL in mobile health applications.
method Advantage learning framework using optimal Q-estimators.
result New policy converges faster than existing methods.
FedLCB-Q learns optimal policies from federated offline data with linear speedup.
problem Learning optimal policies from offline data with federated learning.
method Federated offline RL algorithm tailored for Q-learning, using local Q-function updates and central aggregation.
result Achieves linear speedup in sample complexity with collaboration among agents.
Pipeline selects best RL policy from limited data.
problem Selecting best RL policy from small datasets.
method Task- and method-agnostic pipeline using multiple data splits.
result Pipeline produces higher-performing policies.
Paper proposes a method to optimize policies for diverse individuals using heterogeneous data.
problem Learning optimal policies for a heterogeneous population from pre-collected data.
method Individualized offline policy optimization framework for heterogeneous MDPs.
result The proposed P4L algorithm achieves a fast rate of average regret.
Selective state-adaptive regularization improves offline RL performance.
problem Extrapolation errors and value overestimation in static dataset RL.
method State-adaptive regularization coefficients trust Bellman-driven results selectively.
result Significant improvement in performance on D4RL benchmark.
Enhances reinforcement learning from sparse data.
problem Limited data for offline reinforcement learning.
method Trajectory-based data augmentation.
result Improves reinforcement learning performance.
Proposes RFQI for robust RL using offline data.
problem Learning robust policies in the presence of model uncertainty.
method RFQI algorithm using offline data to learn optimal robust policy.
result RFQI learns near-optimal robust policy under standard assumptions.
MoMA improves model-based RL by using unrestricted policy classes.
problem Limited sample efficiency and generalizability in model-based offline RL.
method Model-based mirror ascent algorithm with general function approximations.
result Theoretical guarantees and practical implementation of MoMA.
Study tackles balancing policy switching costs in offline RL.
problem Balancing the cost of policy switching in offline RL.
method Optimal transport ideas and Net Actor-Critic algorithm.
result Demonstrated efficiency on multiple RL benchmarks.
Oracle-efficient algorithm for offline RL with partial data coverage.
problem Offline reinforcement learning with partial data coverage and constraints.
method PDOCRL, a primal-dual algorithm with decomposed linear-programming formulation.
result Near-optimal, near-feasible policy with \(\widetilde{\mathcal O}(ε^{-2})\) sample guarantee.
New policy for non-stationary multi-armed bandits with recovering rewards.
problem Non-stationary multi-armed bandits with recovering rewards.
method Purely Periodic Policies integrating upper confidence bound procedure.
result Approximately O ~ ( N T ) \widetilde{\mathcal O}(N\sqrt{T}) O ( N T ) regret against the offline benchmark. Paper tackles policy selection in offline RL without hyperparameters.
problem Selecting between policies and value functions in offline RL.
method Designs hyperparameter-free algorithms based on BVFT for policy selection.
result Demonstrates effectiveness in discrete-action benchmarks like Atari.
BOMS enhances offline MBRL by improving model selection with Bayesian optimization.
problem Inaccurate model selection in offline MBRL due to distribution shift.
method Proposes BOMS, an active model selection framework using Bayesian optimization.
result Improves model selection with only a small amount of online interaction.
BREMEN optimizes policies offline with fewer data, achieving efficient deployment.
problem High cost of updating policies in real-world applications.
method Behavior-Regularized Model-ENsemble (BREMEN) algorithm for offline optimization.
result BREMEN achieves impressive deployment efficiency with 5-10 deployments, outperforming standard RL methods.
OSIL learns safe policies from unsafe demonstrations.
problem Offline safe imitation learning with implicit safety.
method Formulates CMDP, infers safety from non-preferred trajectories, learns cost model.
result OSIL learns safer policies without degrading reward performance.
Paper tackles offline RL with weak assumptions on both function classes and data coverage.
problem Achieve sample-efficient offline RL with weak assumptions on both factors.
method Simple algorithm based on primal-dual formulation of MDPs, with density-ratio function modeling dual variables.
result Polynomial sample complexity achieved under realizability and single-policy concentrability.
Simplifies BCQ to match and outperform state-of-the-art in offline RL benchmarks.
problem Sample efficiency in offline reinforcement learning.
method Introduces EMaQ, a novel backup operator for offline RL.
result EMaQ matches and outperforms prior state-of-the-art in offline RL benchmarks.
Paper tackles policy selection with logged data and limited online interactions.
problem Safe evaluation and deployment of offline reinforcement learning policies.
method Active offline policy selection combining logged data with online interaction.
result Improves upon state-of-the-art OPE estimates and pure online policy evaluation.
A new framework for offline RL improves policy flexibility and regularity.
problem Lack of environmental interactions in offline RL leads to poor policy performance.
method Proposes a behavior-regularized implicit policy framework with modified policy-matching methods.
result The framework improves policy effectiveness and robustness beyond static datasets.
Study human-machine interaction with private info using offline RL.
problem Confounding bias and distributional mismatch in offline RL for human-guided interaction.
method Developed a novel identification result and OPE method to address confounding bias, and used pessimism to tackle distributional mismatch.
result Policy pair converges to optimal one at satisfactory rate under mild assumptions.
Off-policy reinforcement learning (RL) using a fixed offline dataset of logged interactions is an important consideration in real world applications. This paper studies offline RL using the DQN replay dataset comprising the entire replay experience of a DQN agent on 60 Atari 2600 games. We demonstrate that recent off-p…