arXiv research

Develops a method to estimate optimal policy value in online learning.

Optimistic NPG improves policy optimization in online RL with efficient sample complexity.

This work explains why online imitation learning improves faster than theory predicts.

This paper bridges offline and online RL by studying policy finetuning with a reference policy.

Transfer learning significantly accelerates the reinforcement learning process by exploiting relevant knowledge from previous experiences. The problem of optimally selecting source policies during the learning process is of great importance yet challenging. There has been little theoretical analysis of this problem. In…

Paper tackles policy selection with logged data and limited online interactions.

RANDomized-exploration policy Optimization via Multiple Importance Sampling with Truncation (RANDOMIST) for PO with mediator feedback.

New control methods for systems with adversarial perturbations.

A new method combines online and offline learning to tackle contextual bandits with missing action support.

This paper explores how imperfect reward models can improve online RLHF.

Paper proposes an $ε$-policy gradient for online pricing, reducing regret to $\mathcal{O}(\sqrt{T})$.

New method tackles safe reinforcement learning from offline data.

New method optimizes policies in non-stationary environments.

New RL algorithm tackles online robust MDPs with uncertainty.

Boosting improves online decision-making for large expert sets.

A3RL combines online and offline RL with active sampling to improve policy learning.

DR-RPO optimizes robust policies in RL with limited interaction, achieving sublinear regret.

The paper introduces SuccessProbaMax to optimize policy success probability in online advertising.

Paper optimizes GAIL for online and offline learning with linear approximations.

Evolutionary Strategies optimize hyper-parameters for off-policy learning.

New algorithm reduces decision switching in dynamic environments.

Paper addresses online alignment of large language models under uncertain preference feedback.

Study non-asymptotic BPI guarantees for online RL.

LF-IBIS learns optimal policies online without explicit likelihood.

Optimal online data collection for semiparametric inference reduces regret.

New $L_1$-Coverage objective simplifies exploration in reinforcement learning.

This paper presents competitive algorithms for a novel class of online optimization problems with memory. We consider a setting where the learner seeks to minimize the sum of a hitting cost and a switching cost that depends on the previous $p$ decisions. This setting generalizes Smoothed Online Convex Optimization. The…

This work proposes robust reinforcement learning methods using both offline and online data.

In this paper, we propose a novel framework for approximating the explicit MPC law for linear parameter-varying systems using supervised learning. In contrast to most existing approaches, we not only learn the control policy, but also a "certificate policy", that allows us to estimate the sub-optimality of the learned …

Algorithm extsc{Pedel} learns near-optimal policies efficiently on specific problems.

Model-based reinforcement learning (MBRL) with model-predictive control or online planning has shown great potential for locomotion control tasks in terms of both sample efficiency and asymptotic performance. Despite their initial successes, the existing planning methods search from candidate sequences randomly generat…

In real-world machine learning applications, there is a cost associated with sampling of different features. Budgeted learning can be used to select which feature-values to acquire from each instance in a dataset, such that the best model is induced under a given constraint. However, this approach is not possible in th…

FRONT optimizes decisions with interference, reducing regret over time.

New policy for non-stationary multi-armed bandits with recovering rewards.

This paper optimizes slate decision systems for large action spaces.

Develops a method to efficiently use offline data for RL policy optimization.

Fine-tuning RL with offline data reduces online interactions.

We consider a learning system based on the conventional multiplicative weight (MW) rule that combines experts' advice to predict a sequence of true outcomes. It is assumed that one of the experts is malicious and aims to impose the maximum loss on the system. The loss of the system is naturally defined to be the aggreg…

We introduce Bayesian least-squares policy iteration (BLSPI), an off-policy, model-free, policy iteration algorithm that uses the Bayesian least-squares temporal-difference (BLSTD) learning algorithm to evaluate policies. An online variant of BLSPI has been also proposed, called randomised BLSPI (RBLSPI), that improves…

New algorithm for online learning in episodic MDPs with convex objectives.

Unified hybrid RL algorithm improves online RL performance with offline data.

New method for efficient online exploration in RLHF reduces regret.

Efficiently identifies best policies in tabular MDPs with reduced computational cost.

We present a predictor-corrector framework, called PicCoLO, that can transform a first-order model-free reinforcement or imitation learning algorithm into a new hybrid method that leverages predictive models to accelerate policy learning. The new "PicCoLOed" algorithm optimizes a policy by recursively repeating two ste…

AWAC combines offline and online data to accelerate RL learning.

In this paper, we develop a multi-agent reinforcement learning (MARL) framework to obtain online power control policies for a large energy harvesting (EH) multiple access channel, when only causal information about the EH process and wireless channel is available. In the proposed framework, we model the online power co…

New method optimizes experiments under constraints.

Paper introduces efficient online sub-sampling for RL with function approximation, reducing policy updates.