New method reduces state distribution mismatch in off-policy RL.
problem State distribution mismatch in off-policy RL algorithms.
method Develops a novel constrained off-policy gradient objective to minimize state distribution shift.
result Minimizing state distribution shift improves performance in off-policy RL algorithms.
New method corrects state distribution mismatch for off-policy policy optimization.
problem Mismatch between behavior and evaluation policy state distributions.
method Off-policy policy gradient with state distribution correction.
result Significantly improved policy quality in simulations.
New method estimates state-action stationary distribution for better off-policy policy evaluation.
problem Accurately estimating state-action stationary distribution for off-policy policy evaluation.
method Estimated Mixture Policy (EMP) for state and state-action stationary distribution corrections.
result Empirical validation shows improved accuracy over state-of-the-art methods.
OSIRIS reduces variance in off-policy evaluation by omitting irrelevant states.
problem High variance in importance sampling-based OPE estimators.
method OSIRIS reduces variance by omitting likelihood ratios associated with states irrelevant to return.
result OSIRIS is unbiased and has lower variance than ordinary importance sampling.
Improves off-policy evaluation weights for balanced state-action pair distribution.
problem Imbalance in importance sampling weights for off-policy evaluation of contextual bandits.
method Balanced Off-Policy Evaluation (B-OPE) method that minimizes imbalance to desired counterfactual distribution of state-action pairs.
result Experimental evidence shows B-OPE improves offline policy evaluation in both discrete and continuous action spaces.
Paper unifies off-policy learning algorithms and introduces C-trace for better trade-offs.
problem Improving efficiency and scalability in off-policy learning.
method Unified view of off-policy algorithms, considering update variance, fixed-point bias, and contraction rate trade-offs.
result C-trace algorithm demonstrates better trade-offs and state-of-the-art performance.
This paper simplifies OPE in large state spaces using state abstractions.
problem Accurately evaluating policies offline in large state spaces.
method Developed a backward-model-irrelevance condition and an iterative state abstraction procedure.
result Deeply-abstracted states substantially simplify OPE sample complexity.
Paper introduces a new value function for state transitions and optimal policy learning.
problem Learning optimal policies from state transitions and actions.
method Develops a forward dynamics model to maximize a novel value function Q(s,s′). result Demonstrates benefits in value function transfer, redundant action spaces, and off-policy learning.
This paper studies optimal approximation factors in misspecified off-policy RL, identifying key factors under various settings.
problem Understanding optimal approximation factors in misspecified off-policy value function estimation.
method Examined various settings including weighted L2-norm, L∞ norm, state aliasing, and state coverage. result Established optimal asymptotic approximation factors for different norms and identified two instance-dependent factors for L2(μ) norm. Improves RL algorithms with two techniques.
problem Enhance off-policy RL performance.
method Formulates RL as proximal point iteration; uses value functions for improved action value estimate.
result Significant performance improvement on RL benchmarks.
New method optimizes policies in non-stationary environments.
problem Optimizing policies in non-stationary, context-dependent environments.
method Two-phase approach: offline learning and online adaptation.
result Our method outperforms existing approaches in both synthetic and real-world datasets.
Deep network improves NILM by distinguishing on/off states of appliances.
problem Break down household aggregate electricity consumption into individual appliance usages.
method Subtask gated network combining regression and classification subtasks.
result Surpasses state-of-the-art performance for most benchmark cases.
Stabilizes policy optimization with off-policy data using divergence augmentation.
problem Premature convergence and instability in policy optimization with off-policy data.
method Incorporates Bregman divergence between behavior and current policies to ensure safe policy updates.
result Empirically shows better performance in data-scarce scenarios compared to other algorithms.
Novel framework proves fast RL convergence in continuous spaces.
problem Analyzing stability in continuous state-action RL.
method Introduces a novel framework to analyze stability properties of RL.
result Highlights two key stability properties and demonstrates their satisfaction in RL.
STAR framework reduces OPE variance by distilling complex problems into discrete ARPs.
problem High variance and bias in off-policy evaluation methods.
method STAR framework that includes various OPE estimators and leverages state abstraction.
result Predictions from ARPs estimated from off-policy data are asymptotically correct.
We propose and analyze an alternate approach to off-policy multi-step temporal difference learning, in which off-policy returns are corrected with the current Q-function in terms of rewards, rather than with the target policy in terms of transition probabilities. We prove that such approximate corrections are sufficien…
Paper introduces a meta-critic for accelerating off-policy actor-critic learning.
problem Improving sample efficiency in continuous control tasks.
method Meta-critic that meta-learns an additional loss for the actor.
result Online meta-critic learning leads to improved performance in various continuous control environments.
Evolutionary Strategies optimize hyper-parameters for off-policy learning.
problem Hyper-parameter sensitivity in off-policy learning.
method Application of Evolutionary Strategies for online hyper-parameter tuning.
result Our method outperforms state-of-the-art baselines.
A streamlined DRL algorithm improves sample efficiency without entropy maximization.
problem Improving sample efficiency in off-policy DRL algorithms.
method Output normalization and non-uniform sampling.
result Proposed algorithm matches SAC's performance without entropy maximization and improves sample efficiency.
The ability to perform effective off-policy learning would revolutionize the process of building better interactive systems, such as search engines and recommendation systems for e-commerce, computational advertising and news. Recent approaches for off-policy evaluation and learning in these settings appear promising. …
RPG improves sample efficiency in RL by learning optimal action ranks.
problem Sample inefficiency in reinforcement learning.
method Ranking Policy Gradient (RPG) method that learns optimal rank of discrete actions.
result RPG reduces sample complexity and improves sample efficiency for large-scale problems.
PBVFs generalize across policies using learned value functions.
problem RL algorithms forget information about old policies when updating value functions to track the learned policy.
method Introduce Parameter-Based Value Functions (PBVFs) that include policy parameters in their inputs, enabling them to generalize across different policies.
result PBVFs enable zero-shot learning of new policies that outperform any policy seen during training.
New algorithm stabilizes RL policy learning through divergence regularization.
problem Stabilize policy learning and improve performance in RL.
method Proximity term constraining discounted state-action visitation distributions to be close to each other.
result Proposed algorithm improves stability and final performance in RL tasks.
For an autonomous agent, executing a poor policy may be costly or even dangerous. For such agents, it is desirable to determine confidence interval lower bounds on the performance of any given policy without executing said policy. Current methods for exact high confidence off-policy evaluation that use importance sampl…
The paper shows how to stabilize off-policy reinforcement learning using specific state representations.
problem Stability issues in reinforcement learning with function approximation and off-policy learning.
method Formal analysis of representation learning schemes based on the transition matrix of a policy.
result Schur and orthogonal bases of the Krylov subspace provide stable representations for TD learning.
Paper introduces control variate to reduce variance in off-policy reinforcement learning.
problem High variance in off-policy reinforcement learning.
method Introduces control variate technique to Expected Sarsa(λ) and proposes ES(λ)-CV algorithm.
result Proposed ES(λ)-CV algorithm has lower variance than Expected Sarsa(λ).
Study off-policy evaluation and learning in dynamic pricing with context.
problem Dynamic personalized pricing and operations management problems with high-dimensional user types.
method Formalize causal structure, leverage single time-step evaluation, estimate marginal MDP.
result Improved out-of-sample policy performance in dynamic and capacitated pricing.
This paper reviews off-policy evaluation methods in reinforcement learning.
problem Efficiency and accuracy of off-policy evaluation methods in reinforcement learning.
method Discussion of existing OPE methods, their statistical properties, and related research directions.
result Efficiency bounds and state-of-the-art OPE methods in reinforcement learning.
P3O merges on-policy and off-policy updates without extra hyper-parameters.
problem Combining on-policy and off-policy RL algorithms to reduce sample complexity.
method Interleaves off-policy updates with on-policy updates using effective sample size.
result P3O reduces sample complexity of state-of-the-art algorithms.
A new approach to reinforcement learning by adjusting stepsize based on state criticality.
problem The bias-variance trade-off in n-step reinforcement learning algorithms.
method Adapting the stepsize (n) for each state based on a human trainer's criticality measure.
result Improved reinforcement learning performance by dynamically adjusting the stepsize.
Study tackles OPE in confounded settings, estimating policy value from proxies.
problem Difficulty in OPE due to unobserved confounders in infinite-horizon RL.
method Two-stage approach: estimating stationary distribution ratios and combining optimal balancing.
result Policy value can be identified from off-policy data with proxies and latent variable model.
New algorithm for sequential off-policy learning improves performance over batch methods.
problem Training policies from logged interaction data in a sequential setting.
method Combines Logarithmic Smoothing with online PAC-Bayesian tools.
result Improves performance and accelerates convergence in sequential off-policy learning.
Combines parametric and nonparametric models for better off-policy evaluation.
problem Improving off-policy evaluation in reinforcement learning.
method Mixture-of-experts approach combining parametric and nonparametric models.
result Mixture-based approach outperforms individual models and state-of-the-art estimators.
New method for off-policy evaluation in contextual bandits reduces bias-variance tradeoff.
problem Improving off-policy evaluation in contextual bandits.
method Asymptotically optimal doubly robust estimator with shrinkage of importance weights.
result New estimators outperform state-of-the-art methods in experiments.
New RL approach uses future state and action visitation measures for better exploration.
problem Improving exploration in reinforcement learning.
method Intrinsic reward based on future state and action visitation measures, using contraction operators.
result Policies achieve good state-action space coverage and high performance.
Combines pseudo-point and state space approximations for scalable GPs.
problem Handling large numbers of off-the-grid spatial data-points and long time-series.
method Combines pseudo-point approximations for spatial data with state space GP approximations for temporal data.
result Combined approach is more scalable and applicable to a greater range of spatio-temporal problems.
Meta-Q-Learning improves meta-Reinforcement Learning using past data.
problem Improving meta-Reinforcement Learning performance.
method Meta-Q-Learning combines Q-learning, multi-task objectives, and off-policy updates to adapt policies from past data.
result Meta-Q-Learning compares favorably with state-of-the-art meta-RL algorithms on benchmarks.
Unified DICE estimators as regularized Lagrangians for improved off-policy evaluation.
problem Improving off-policy evaluation from behavior-agnostic data.
method Unified derivation of DICE estimators as regularized Lagrangians of a linear program.
result Dual solutions offer greater flexibility and provide superior estimates in practice.
New expressive losses improve adversarial robustness without sacrificing accuracy.
problem Training networks for robustness at the expense of accuracy.
method Formalizing expressivity, using convex combinations of adversarial attacks and IBP bounds.
result Trivial expressive losses yield state-of-the-art results in various settings.
VLBM learns MDP transitions from limited data, improving OPE performance.
problem Limited coverage of state and action space in offline trajectories.
method VLBM uses variational inference with RSA and branching architecture.
result VLBM outperforms existing OPE methods on deep OPE benchmark.
Study on future-dependent value functions for off-policy evaluation in complex environments.
problem Exponential dependence on horizon in off-policy evaluation for complex observations.
method Developed novel coverage assumptions for POMDPs to achieve polynomial bounds.
result Achieved polynomial bounds on previously exponential quantities, improving off-policy evaluation.
Paper proposes a learning-based sparse Bayesian method for accurate off-grid DOA estimation.
problem One-bit off-grid direction of arrival (DOA) estimation in a single snapshot scenario.
method Formulated off-grid DOA estimation model, used Sparse Bayesian framework, proposed Learning-based Sparse Bayesian approach.
result Improved computational efficiency and accuracy in off-grid DOA estimation.
Off-policy evaluation for MNAR rewards in MDPs
problem Off-policy evaluation in MDPs with MNAR rewards
method Formalizing a reward-dependent propensity model and using future states as shadow variables
result Proposed an Fitted-Q-Evaluation-style estimator that propagates recovered rewards while allowing target policies to depend on past missingness indicators
State2vec improves RL by learning state embeddings that generalize across policies.
problem Inefficient generalization across policies in RL.
method Extends node2vec to learn state embeddings accounting for discounted future state transitions.
result Captures the geometry of the state space, leading to sample-efficient value function approximation.
Deep RL agents perform poorly on unseen states, challenging the notion of generalization.
problem Challenges in understanding and measuring generalization in deep reinforcement learning.
method Proposed definitions of generalization, practical evaluation methods, and demonstrations on a benchmark task.
result Deep Q-networks learn poor representations for states not selected on-policy, even if they are not adversarial.
Pareto MTL finds optimal solutions for multiple tasks with different trade-offs.
problem Finding a single optimal solution for multiple conflicting tasks.
method Formulate multi-task learning as multiobjective optimization, decompose into subproblems, solve in parallel.
result Generates well-representative Pareto optimal solutions for different trade-offs.
New methods estimate policy value and gradients for deterministic policies from off-policy data.
problem Estimating policy value and gradients for deterministic policies from off-policy data.
method Proposed new doubly robust estimators based on kernelization approaches.
result Demonstrated a rate independent of horizon length for policy value and gradient estimation.
PRL improves off-policy evaluation in partially observed MDPs.
problem Confounding and bias in offline reinforcement learning with unobserved state factors.
method Extends proximal causal inference to POMDPs, identifying and estimating target policy value.
result Semiparametrically efficient estimators for PRL in partially observed MDPs.