New method improves consistency of reinforcement learning performance evaluations.
problem Inconsistent performance results in reinforcement learning due to flawed evaluation metrics.
method Proposes a new comprehensive evaluation methodology for reinforcement learning algorithms.
result Demonstrates improved reliability of performance measurements for reinforcement learning algorithms.
Paper develops robust policy evaluation for reinforcement learning with outlier and heavy-tailed rewards.
problem Outlier contamination and heavy-tailed rewards in reinforcement learning.
method Develops a fully online robust policy evaluation procedure and efficient statistical inference.
result Establishes the Bahadur-type representation of the estimator and develops an online inference procedure.
New method extends fitted Q-evaluation for distributional off-policy reinforcement learning.
problem Estimating return distribution in reinforcement learning using offline data.
method Developed a set of guiding principles and new FDE methods with theoretical justification.
result FDE methods outperform existing approaches in simulations and real-world games.
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.
Paper proposes a framework for reliable off-policy evaluation in reinforcement learning.
problem Quantifying uncertainty in off-policy estimates for safe deployment of target policies.
method Distributionally robust optimization for creating confidence bounds.
result Non-asymptotic and asymptotic guarantees for robust cumulative reward estimates.
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.
Proposes a method to estimate policy values in reinforcement learning with unmeasured confounders.
problem Estimating policy values in reinforcement learning with unmeasured confounders.
method Develops a two-way deconfounder algorithm using a neural tensor network to learn unmeasured confounders and system dynamics.
result Consistent policy value estimation through model-based estimator.
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.
Study on meta-reinforcement learning generalization in high-dimensional tasks.
problem Generalization performance of meta-reinforcement learning algorithms in high-dimensional tasks.
method High-dimensional, procedurally generated environments.
result Meta-reinforcement learning algorithms exhibit strong overfitting on challenging tasks.
Deep reinforcement learning algorithms have shown an impressive ability to learn complex control policies in high-dimensional tasks. However, despite the ever-increasing performance on popular benchmarks, policies learned by deep reinforcement learning algorithms can struggle to generalize when evaluated in remarkably …
Paper improves bootstrapping for off-policy reinforcement learning inference.
problem Improving bootstrapping for off-policy reinforcement learning inference.
method Proposes a bootstrapping FQE method for off-policy statistical inference and a subsampling procedure to improve runtime.
result Asymptotically efficient and distributionally consistent bootstrapping FQE method for off-policy inference.
Backward exploration reduces sample complexity in policy evaluation.
problem Empirical policy evaluation in reinforcement learning.
method Backward exploration algorithms from high-cost states.
result Reduced average-case sample complexity to O(logS). Deep reinforcement learning enables algorithms to learn complex behavior, deal with continuous action spaces and find good strategies in environments with high dimensional state spaces. With deep reinforcement learning being an active area of research and many concurrent inventions, we decided to focus on a relatively …
New algorithms improve policy evaluation in reinforcement learning.
problem Off-policy stability and on-policy efficiency issues in policy evaluation.
method Introduced novel algorithms using oblique projection method.
result Demonstrated both off-policy stability and on-policy efficiency.
Deep RL approach improves MIS for complex environments.
problem Improving off-policy evaluation for complex environments.
method Uses successor representation from deep RL to decouple reward and dynamics.
result Empirically stable and applicable to high-dimensional domains.
A new framework uses multi-agent reinforcement learning for evaluating policies in two-sided markets.
problem Evaluating the effects of different policies in two-sided markets with spatial and temporal interference.
method Introduces a multi-agent reinforcement learning (MARL) framework to address policy evaluation challenges in large-scale fleet management.
result Proposes novel estimators for mean outcomes under different products that are consistent despite high-dimensionality.
Method selects best estimator for off-policy evaluation.
problem Choosing the best estimator for off-policy evaluation.
method Generic data-driven method for estimator selection.
result Method is competitive with oracle estimator, up to a constant factor.
Paper develops a method to highlight influential transitions for off-policy evaluation.
problem Validating off-policy evaluation in high-stakes settings.
method Highlight influential transitions, develop influence functions for Q-evaluation.
result Method can identify limitations and make evaluation more robust.
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.
Off-policy evaluation (OPE) in reinforcement learning allows one to evaluate novel decision policies without needing to conduct exploration, which is often costly or otherwise infeasible. We consider for the first time the semiparametric efficiency limits of OPE in Markov decision processes (MDPs), where actions, rewar…
Meta-reinforcement learning improves fault-adaptive control efficiency.
problem Adaptive control under abrupt system faults with strict time constraints.
method Model-agnostic meta learning (MAML) with a fault library of prior policies.
result Improved sample efficiency and quick adaptation to new faults.
PRUDEX-Compass evaluates FinRL methods on 6 axes for financial market investments.
problem Insufficient evaluation of FinRL methods in financial markets.
method Introduces PRUDEX-Compass with 6 axes and 17 measures for evaluation.
result Demonstrates the effectiveness of PRUDEX-Compass on 4 real-world datasets.
Paper introduces DQPOPE for estimating return distributions in reinforcement learning.
problem Estimating the entire return distribution from off-policy data.
method Deep quantile process regression for distributional off-policy evaluation.
result DQPOPE achieves statistical advantages by estimating full return distribution with same sample size.
Evaluation of deep reinforcement learning (RL) is inherently challenging. In particular, learned policies are largely opaque, and hypotheses about the behavior of deep RL agents are difficult to test in black-box environments. Considerable effort has gone into addressing opacity, but almost no effort has been devoted t…
Self-consistent models improve reinforcement learning by aligning predictions with future values.
problem Improving reinforcement learning by aligning model predictions with future values.
method Proposes multiple self-consistency updates to encourage a learned model and value function to be consistent with each other.
result Self-consistency helps both policy evaluation and control in both tabular and function approximation settings.
Paper establishes baselines for offline RL from visual observations.
problem Challenges in offline reinforcement learning from visual observations with continuous action spaces.
method Simple baselines and benchmarking tasks for offline RL from visual observations.
result Simple modifications to existing online RL algorithms outperform existing offline RL methods.
A deep reinforcement learning approach for slate re-ranking in e-commerce.
problem Improving user satisfaction in e-commerce by optimizing the ranking of items in a slate.
method Generator and Critic approach, using reinforcement learning and a Full Slate Critic model.
result The Generator and Critic approach significantly outperforms existing methods in slate evaluation and efficiency.
Reinforcement learning is well suited for optimizing policies of recommender systems. Current solutions mostly focus on model-free approaches, which require frequent interactions with the real environment, and thus are expensive in model learning. Offline evaluation methods, such as importance sampling, can alleviate s…
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.
SAMBA improves safe reinforcement learning with active exploration metrics.
problem Safe reinforcement learning in dynamic systems.
method Combines probabilistic modelling, information theory, and statistics. Uses novel metrics for out-of-sample Gaussian process evaluation.
result Orders of magnitude reduction in samples and violations compared to state-of-the-art methods.
Deep RL for portfolio management shows poor robustness.
problem Robustness of Deep RL algorithms in online portfolio management.
method Proposed a training and evaluation process for assessing DRL algorithms.
result Most Deep RL algorithms are not robust, generalizing poorly and degrading quickly.
This paper optimizes off-policy evaluation in reinforcement learning with function approximation.
problem Estimating cumulative value of a new policy from logged data generated by an unknown policy.
method Regression-based fitted Q iteration method, equivalent to estimating conditional mean embedding of transition operator.
result The method is minimax-optimal, with nearly minimal estimation error.
A novel fully asynchronous scheme for distributed reinforcement learning over networks.
problem Policy evaluation in distributed reinforcement learning over networks.
method Design of a stochastic average gradient (SAG) based distributed algorithm and push-pull augmented graph approach.
result The proposed algorithm converges at a linear rate of \(\mathcal{O}(c^k)\) with \(c\in(0,1)\) and \(k\) increasing by one per node update.
Paper introduces a new distributional successor measure for reinforcement learning.
problem Learning the distributional consequences of behavior in reinforcement learning.
method Formulates distributional successor measure as a distribution over distributions, proposes algorithm to learn it from data.
result Demonstrates zero-shot risk-sensitive policy evaluation.
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.
New method reduces sample complexity for robust reinforcement learning.
problem Finite sample analysis in robust reinforcement learning.
method Stochastic approximation framework with controlled bias, using MLMC techniques and geometric truncation.
result Order-optimal sample complexity of ildeO(ε−2) for robust policy evaluation. This paper uses CausalGANs and RL with LLM to predict bond yields.
problem Challenges in financial bond yield forecasting due to data scarcity and market conditions.
method Proposes a novel framework combining CausalGANs, RL, and LLM for synthetic data generation and trading signals.
result Improves forecasting performance over existing methods with low Mean Absolute Error.
This work benchmarks MARL algorithms in cooperative tasks.
problem Lack of evaluation tasks and criteria for comparing MARL algorithms.
method Systematic evaluation of three MARL algorithm classes in diverse cooperative tasks.
result Insights into the effectiveness of different learning approaches.
Paper improves off-policy evaluation for reinforcement learning with asymptotically efficient estimators.
problem Estimating target policy performance using offline data collected by a different policy.
method Developed a modified marginalized importance sampling (MIS) estimator that achieves asymptotically efficient error bounds.
result Proved that a simple modification to the MIS estimator can achieve a Cramer-Rao lower bound in mean square error.
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.
New algorithms for model selection in off-policy evaluation of reinforcement learning.
problem Hyperparameter tuning for off-policy evaluation methods in reinforcement learning.
method Developed new model-free and model-based selectors with theoretical guarantees and a new experimental protocol.
result New model-free selector, LSTD-Tournament, demonstrates promising empirical performance.
Pretraining reinforcement learning methods with demonstrations has been an important concept in the study of reinforcement learning since a large amount of computing power is spent on online simulations with existing reinforcement learning algorithms. Pretraining reinforcement learning remains a significant challenge i…
This paper unifies three regularization methods in batch reinforcement learning.
problem Learning overly-complex models in batch reinforcement learning.
method Unified weighted average transition matrix framework for three regularization methods.
result Empirical evaluation confirms intuitions about regularization methods' performance.
A new method estimates multi-dimensional value distributions using Hilbert space embeddings.
problem Estimating value distributions in complex, multi-dimensional reinforcement learning settings.
method Hilbert space mappings and kernel mean embeddings to estimate the kernel mean embedding of multi-dimensional value distributions.
result Uniform convergence guarantees and robust off-policy evaluation demonstrated in simulations.
One of the key challenges in applying reinforcement learning to real-life problems is that the amount of train-and-error required to learn a good policy increases drastically as the task becomes complex. One potential solution to this problem is to combine reinforcement learning with automated symbol planning and utili…
Meta-reinforcement learning algorithms can enable robots to acquire new skills much more quickly, by leveraging prior experience to learn how to learn. However, much of the current research on meta-reinforcement learning focuses on task distributions that are very narrow. For example, a commonly used meta-reinforcement…
The paper analyzes reinforcement learning methods for estimating weights and quality functions with fast convergence rates.
problem Estimating weights and quality functions in reinforcement learning with function approximation.
method The paper uses minimax methods for estimating marginal importance weights and q-functions.
result The minimax approach enables fast rates of convergence for weights and quality functions, achieving first-order efficiency.
QLBS and RLOP methods improve option pricing and hedging performance.
problem Improving option pricing and hedging performance under market frictions.
method Incorporates risk aversion and trading costs into QLBS, proposes RLOP approach.
result RLOP outperforms in dynamic hedging by reducing shortfall probability.