Policy-gradient method controls multiple non-cohesive targets.
problem Controlling multiple non-cohesive targets in a decentralized manner.
method Proximal Policy Optimization for target selection and driving.
result Effective control of non-cohesive targets without prior dynamics knowledge.
Framework improves policy generalizability under biased training data.
problem Learning policies that generalize to a target population from biased training data.
method Characterizes sample selection bias using a selection variable, optimizes minimax value over uncertainty set, derives efficient algorithm.
result Policies generalize to target population, outperform standard methods.
Study designs logging policies to minimize off-policy evaluation error.
problem Minimizing OPE error with logging policies for target policies.
method Characterizes reward-coverage tradeoff, proposes a unifying framework, derives optimal policies.
result Provides actionable guidance for firms choosing recommendation systems.
We observe that several existing policy gradient methods (such as vanilla policy gradient, PPO, A2C) may suffer from overly large gradients when the current policy is close to deterministic (even in some very simple environments), leading to an unstable training process. To address this issue, we propose a new method, …
In this paper, we present a new approach to Transfer Learning (TL) in Reinforcement Learning (RL) for cross-domain tasks. Many of the available techniques approach the transfer architecture as a method of speeding up the target task learning. We propose to adapt and reuse the mapped source task optimal-policy directly …
The paper targets optimal interventions for long-term outcomes using imputed data and policy learning.
problem Maximizing long-term outcomes observed only in the future.
method Imputing missing long-term outcomes and using a doubly-robust approach for policy evaluation and optimization.
result The approach outperforms simple short-term proxies and achieves significant revenue impact over three years.
Off-policy learning exhibits greater instability when compared to on-policy learning in reinforcement learning (RL). The difference in probability distribution between the target policy (π) and the behavior policy (b) is a major cause of instability. High variance also originates from distributional mismatch. The var…
RL policy tracks dynamic targets in partially known environments robustly.
problem Active target tracking in partially known environments.
method Deep reinforcement learning (RL) approach for in-sight tracking, navigation, and exploration.
result Unified RL policy shows robust behavior for agile and anomalous targets.
Proposes a method to improve treatment policies in data-scarce clinical settings.
problem Improving treatment policies in data-scarce clinical settings with unobserved confounding.
method Uses a causal mechanism to model the underlying generative process and augments counterfactual trajectories with source domain priors.
result Significantly improves treatment policy performance in a simulated sepsis treatment task.
Estimates personalized policies robust to shifts in target populations.
problem Estimating policies that perform well in diverse target populations.
method Develops methods for estimating robust policies considering shifts in outcomes and characteristics.
result Welfare-maximizing policies are robust to certain shifts in potential outcomes.
New model-based methods adapt pre-trained policies to unseen environments efficiently.
problem High sample complexity in reinforcement learning limits practical applications.
method Combines online learning and adaptive control to adapt policies in unseen environments.
result Proves policies can quickly recover trajectories from source to target environments.
Develops methods to estimate and quantify uncertainty in off-policy evaluation.
problem Uncertainty quantification in off-policy evaluation for new policy deployment.
method Designs a pseudo policy to generate subsamples and applies conformal prediction.
result Valid interval estimators for target policy's return with uncertainty quantification.
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…
PTF accelerates RL by reusing source policies without measuring task similarity.
problem Leveraging prior knowledge for faster RL.
method Adaptive Policy Transfer Framework (PTF) for RL.
result Significantly accelerates RL learning process and surpasses state-of-the-art methods.
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.
Combines trial and observational data to improve policy evaluation.
problem External validity of randomized trial results in target populations.
method Uses covariate data to model trial sampling and certifies policy evaluations.
result Valid trial-based policy evaluations under model miscalibration.
Paper proposes a method to create more reliable confidence intervals for off-policy evaluations.
problem Creating reliable confidence intervals for off-policy evaluations.
method Proposes a deeply-debiasing procedure to construct efficient, robust, and flexible confidence intervals.
result Validated by theoretical results and numerical experiments, the method improves the reliability of off-policy evaluations.
New method stabilizes FQE by reweighting Bellman targets.
problem Stability guarantees for FQE often rely on Bellman completeness, which can fail with function approximation.
method Proposes stationary-weighted FQE, reweighting Bellman targets by stationary target-to-behavior density ratio.
result Proves finite-sample linear convergence to stationary projected Bellman fixed point without Bellman completeness.
RL approach for target tracking with unknown dynamics and sensor control.
problem Tracking an unknown target with sensor control.
method Track-MDP formulation for RL, compared with POMDP.
result Optimal RL policy tracks all target paths with certainty.
Presents SPEED, an algorithm for optimal policy evaluation in linear bandits with heteroscedastic noise.
problem Optimal data collection for policy evaluation in linear bandits with heteroscedastic reward noise.
method Formulated an optimal design for weighted least squares estimates, derived the optimal sample allocation, introduced SPEED algorithm, and derived regret bounds.
result SPEED leads to policy evaluation with MSE comparable to oracle strategy and significantly lower than random policy execution.
The paper tackles fair policy targeting by optimizing allocation rules to minimize unfairness.
problem Discrimination in individualized treatments of social welfare programs.
method Formulated as a mixed-integer linear program, solved using off-the-shelf algorithms, derived regret bounds and small sample guarantees.
result Designs fair and efficient treatment allocation rules within the Pareto frontier.
Scarce data is a major challenge to scaling robot learning to truly complex tasks, as we need to generalize locally learned policies over different task contexts. Contextual policy search offers data-efficient learning and generalization by explicitly conditioning the policy on a parametric context space. In this paper…
Scarce data is a major challenge to scaling robot learning to truly complex tasks, as we need to generalize locally learned policies over different "contexts". Bayesian optimization approaches to contextual policy search (CPS) offer data-efficient policy learning that generalize over a context space. We propose to impr…
Domain randomization (DR) is a successful technique for learning robust policies for robot systems, when the dynamics of the target robot system are unknown. The success of policies trained with domain randomization however, is highly dependent on the correct selection of the randomization distribution. The majority of…
New split rules improve subpopulation targeting in policy-making.
problem Improving binary classification for subpopulation targeting in policy-making.
method MDFS, PFS, wEFS for maximizing distance and penalizing final splits.
result Proposed methods target more vulnerable subpopulations than classic CART/KD-CART.
Data augmentation (DA) has been widely utilized to improve generalization in training deep neural networks. Recently, human-designed data augmentation has been gradually replaced by automatically learned augmentation policy. Through finding the best policy in well-designed search space of data augmentation, AutoAugment…
A drone catches another agile drone using competitive reinforcement learning.
problem Intercepting an agile drone with another agile drone.
method Formulated as a Competitive Reinforcement Learning problem, trained with PPO, using a high-fidelity simulation environment.
result Trained policies outperform common heuristic baselines in catch rate, time to catch, and crash rate.
Generically learns movement control policies from exploration data.
problem Movement optimization in physically based characters.
method Parameterizes actions as target states, learns low-level control policy.
result Improves movement optimization across multiple tasks and algorithms.
New method for estimating value of optimal policies in uncertain scenarios.
problem Inference for optimal policies when they are non-unique or nearly deterministic.
method Semiparametric efficiency bound, uniformly weighted estimator, NSAVE method.
result Proposes NSAVE method for robust inference in uncertain optimal policies.
Learning the value function of a given policy (target policy) from the data samples obtained from a different policy (behavior policy) is an important problem in Reinforcement Learning (RL). This problem is studied under the setting of off-policy prediction. Temporal Difference (TD) learning algorithms are a popular cl…
Study shows how macroprudential policies affect credit growth in Israel, especially in housing and business sectors.
problem Impact of macroprudential policies on credit growth in Israel.
method Bank-level panel data analysis for Israel, 2004-2019; interaction of monetary and macroprudential policies.
result Accommodative monetary policy interacts with macroprudential policies to increase total credit growth.
We study the problem of off-policy evaluation (OPE) in Reinforcement Learning (RL), where the aim is to estimate the performance of a new policy given historical data that may have been generated by a different policy, or policies. In particular, we introduce a novel doubly-robust estimator for the OPE problem in RL, b…
This paper bridges the gap between theoretical and practical OPE for bandit problems.
problem Estimating the value of a target policy from samples generated by different policies.
method Categorizing OPE situations based on evaluation policy properties, proposing a meta-algorithm.
result Meta-algorithm successfully bridges the gap between theoretical and practical OPE for bandit problems.
Paper improves policy updates in reinforcement learning to speed up learning.
problem Slow learning and unlearning in policy optimization.
method Introduces a novel gradient update and a modified policy update.
result Proves modified policy update converges to global optimality.
Recent policy optimization approaches (Schulman et al., 2015a; 2017) have achieved substantial empirical successes by constructing new proxy optimization objectives. These proxy objectives allow stable and low variance policy learning, but require small policy updates to ensure that the proxy objective remains an accur…
DDO-RM improves reward-based policies by converting reward scores into a target distribution.
problem Improving reward-based policies when the reward function is simpler than the policy.
method Converts reward scores into a target distribution and uses KL-regularized mirror-descent updates.
result DDO-RM outperforms DPO in pair accuracy and mean margin.
Transfer reinforcement learning (RL) aims at improving the learning efficiency of an agent by exploiting knowledge from other source agents trained on relevant tasks. However, it remains challenging to transfer knowledge between different environmental dynamics without having access to the source environments. In this …
The paper discusses the role of monetary policy when potential output depends on the inflation rate. If the intention of the central bank is to maximize actual output growth, then it has to be credibly committed to a strict inflation targeting rule, and to take the MOGIR (the Maximizing Output Growth Inflation Rate) as…
A new method for evaluating and selecting policies in contextual bandits improves confidence intervals and policy quality.
problem Evaluating and selecting policies in contextual bandits with logged data.
method Self-normalized Importance Weighting (SN) estimator with Efron-Stein tail inequality and multiplicative bias control.
result The method provides tighter confidence intervals and better policy selection compared to competitors.
The paper analyzes the sample complexities for policy evaluation with linear function approximation.
problem Policy evaluation with linear function approximation in discounted infinite horizon Markov decision processes.
method Investigates sample complexities for two policy evaluation algorithms: TD and TDC.
result Establishes high-probability sample complexity bounds for policy evaluation algorithms.
VA-OPE improves OPE by incorporating variance information, achieving tighter error bounds.
problem Estimating value function of a target policy from offline data collected by a behavior policy.
method Proposes VA-OPE, an algorithm that reweights Bellman residual using estimated variance of the value function.
result Achieves a tighter error bound than the best-known result.
Attackers can poison environments to force RL agents to follow target policies.
problem Security threat to reinforcement learning where attackers manipulate environments to force agents into following target policies.
method Optimization framework for finding optimal stealthy attacks under different measures of attack cost.
result Attackers can easily succeed in teaching any target policy to RL agents under mild conditions.
This paper studies the problem of optimally allocating treatments in the presence of spillover effects, using information from a (quasi-)experiment. I introduce a method that maximizes the sample analog of average social welfare when spillovers occur. I construct semi-parametric welfare estimators with known and unknow…
Computer simulation provides an automatic and safe way for training robotic control policies to achieve complex tasks such as locomotion. However, a policy trained in simulation usually does not transfer directly to the real hardware due to the differences between the two environments. Transfer learning using domain ra…
Paper develops IV method for consistent OPE in confounded MDPs.
problem Confounding variables in observational data affect OPE effectiveness.
method Instrumental variable approach for consistent off-policy evaluation.
result IV method enables correct identification of target policy's value.
WAPPO optimizes feature distributions for better visual transfer in RL.
problem Improving visual transfer in reinforcement learning.
method WAPPO uses Wasserstein Confusion to minimize feature distribution distance.
result WAPPO outperforms previous methods in visual transfer across different environments.
The problem of on-line off-policy evaluation (OPE) has been actively studied in the last decade due to its importance both as a stand-alone problem and as a module in a policy improvement scheme. However, most Temporal Difference (TD) based solutions ignore the discrepancy between the stationary distribution of the beh…
Dual-structured method improves cross-domain imitation learning.
problem Learning policies in a target domain from source domain expert demonstrations.
method Extracts domain and policy expertness features, synthesizes new features.
result Superior performance compared to other algorithms in MuJoCo tasks.