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.
Self-supervised policy adapts after deployment without rewards.
problem Generalizing reinforcement learning policies across different environments.
method Uses self-supervision to train policies in new environments without reward signals.
result Significant improvements in generalization across diverse environments.
New RL algorithm reduces deployment cost for linear function approximations.
problem Efficiently deploying new policies in RL with unknown rewards.
method Proposes an algorithm that minimizes trajectories needed for identifying optimal policies.
result Achieves optimal deployment complexity and sample complexity.
Paper establishes lower bounds and optimal algorithms for deployment-efficient RL.
problem Deployment efficiency in reinforcement learning.
method Optimization with constraints, lower bounds, algorithms.
result Established optimal algorithms for deployment-efficient RL.
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.
Robust OPE framework uses human inputs to improve policy evaluation in changing environments.
problem Inaccurate policy evaluations due to shifts in environment properties.
method Adapts OPE methods to shifts on user-inputted covariates, providing more realistic utility estimates.
result Robust OPE framework yields less pessimistic policy evaluations and captures realistic dataset shifts.
The paper proposes an iterative approach to batch reinforcement learning for safer and more informative data collection.
problem Learning policies that are too rigid and do not adapt to new data.
method Safe diversified model-based policy search in an iterative batch reinforcement learning framework.
result Improved learned policies through continuous data collection and adaptation.
Optimal fund deployment strategy under uncertain deal arrivals.
problem Deciding when to invest in deals with uncertain future arrivals.
method Formulated as CTMDP, solved via ADP with QMC sampling.
result Developed interpretable acceptance policy outperforming baseline.
Improves off-policy evaluation with imperfect annotations.
problem Limited dataset coverage for evaluating new policies.
method Doubly robust estimators combining IS and DM, incorporating counterfactual annotations.
result Using annotations within the DM component yields the most desirable theoretical results.
New RL algorithms improve control tasks with data reuse.
problem Real-world control requires performance guarantees and data efficiency.
method Generalized Policy Improvement combining on-policy guarantees and sample reuse.
result Extensive experimental analysis shows benefits of new algorithms.
Perry uses auxiliary data to estimate RL policy values with confidence intervals.
problem Leveraging auxiliary datasets for off-policy evaluation with uncertainty quantification.
method Two methods for constructing valid confidence intervals for OPE with data augmentation.
result Methods consistently produce confidence intervals covering ground truth policy values.
A new method predicts stock ranking uncertainty to improve trading performance during regime shifts.
problem Ranking models fail during regime shifts, leading to suboptimal performance.
method Adapting DEUP to rankers, predicting rank displacement and uncertainty, and proposing a two-level deployment policy.
result The two-level deployment policy improves risk-adjusted performance and indicates DEUP adds value mainly as a tail-risk guard.
Simple policy outperforms complex ones in cloud auto-scaling.
problem Predicting resource scaling for large-scale cloud applications with limited deployment throughput.
method Probabilistic workload forecast for auto-scaling decisions based on risk aversion.
result The proposed policy outperforms sophisticated and simple benchmark policies in real-world and synthetic data.
Develops confidence bounds for off-policy evaluation in contextual bandits.
problem Evaluating policies that were not used to collect data.
method Martingale analysis for non-asymptotic, non-parametric, and valid confidence sequences.
result Empirically tight bounds on failure probability and width.
Optimizes electric aircraft deployment for Canadian aviation to reduce emissions.
problem Limited fleet capacity and operational structure hinder electric aircraft transition.
method Multi-period mixed-integer linear programming (MILP) framework.
result Electric aircraft can reduce emissions by over 70% within five years.
Improved deep learning model deployment on tiny MCUs with mixed-precision quantization.
problem Memory limitations prevent accurate deployment of DNN models on tiny MCUs.
method Automated mixed-precision quantization using Reinforcement Learning for MCU constraints.
result Mixed-precision models achieve high accuracy with uniform quantization policies.
New method optimizes language model performance for test-time strategies.
problem Mismatch between training objectives and test-time deployment of large language models.
method Tail-Extrapolated estimators to approximate best-of-N performance from limited training rollouts.
result Improved performance of best-of-N deployment across various models and datasets.
Offline RL tackles resource-constrained online deployment with improved policy transfer.
problem Training policies with limited online features using a rich offline dataset.
method Introduce a policy transfer algorithm that first trains a teacher agent with full offline features and then transfers knowledge to a student agent with limited online features.
result Consistent improvement in performance over baseline methods on resource-constrained datasets.
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.
CSA fills a gap in RLVR-trained LLM deployment by providing anytime-valid selective risk control.
problem Deployment of RLVR-trained LLMs in regulated organizations requires a safety certificate for every round without waiting for long-run averages.
method CSA uses a (test statistic, validity guarantee, deployment rule) framework to fill the gap, maintaining a Ville-type e-process per threshold on a Bonferroni grid.
result CSA provides the first anytime-valid selective risk control for RLVR-trained LLMs, matching the long-run average certification rate and satisfying pathwise validity and non-refusing deployment on every cell.
We use deep reinforcement learning to optimize experimental designs efficiently.
problem Optimizing sequential experimental designs with limited exploration and black-box models.
method Reduced the optimal design problem to an MDP and solved it with deep reinforcement learning.
result Our approach achieves state-of-the-art performance on both continuous and discrete design spaces.
We study the problem of designing AI agents that can robustly cooperate with people in human-machine partnerships. Our work is inspired by real-life scenarios in which an AI agent, e.g., a virtual assistant, has to cooperate with new users after its deployment. We model this problem via a parametric MDP framework where…
With the advent of the Internet of Things (IoT), an increasing number of energy harvesting methods are being used to supplement or supplant battery based sensors. Energy harvesting sensors need to be configured according to the application, hardware, and environmental conditions to maximize their usefulness. As of toda…
Proposes a conservative exploration method for RL agents.
problem Guaranteeing performance of exploratory policies in RL.
method Importance sampling for off-policy policy evaluation.
result Derives a regret bound ensuring no conservative constraint violation.
Policy gradient is a generic and flexible reinforcement learning approach that generally enjoys simplicity in analysis, implementation, and deployment. In the last few decades, this approach has been extensively advanced for fully observable environments. In this paper, we generalize a variety of these advances to part…
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.
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.
New framework for learning policies that converge in out-of-sample regions.
problem Reliable out-of-sample recovery in imitation learning.
method Contractive dynamical systems and recurrent equilibrium networks.
result Policy rollouts converge regardless of perturbations, enabling efficient OOS recovery.
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.
PEOC uses policy entropy to detect untrained states in RL.
problem Detecting untrained states in reinforcement learning for safety.
method Policy entropy based one-class classifier.
result PEOC is highly competitive and reliable.
Adapts safe policies for exploration in high-risk settings.
problem Balancing safety and exploration in high-risk environments.
method Uses conformal calibration on a safe reference policy to determine aggressive action limits.
result Safe exploration improves performance without requiring model class identification or hyperparameter tuning.
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.
New method prevents RLHF alignment collapse by accounting for policy's influence on reward model updates.
problem Iterative RLHF leads to alignment collapse where policies exploit RM's blind spots.
method Foresighted policy optimization (FPO) restores missing steering term via regularization.
result FPO prevents alignment collapse on LLM alignment pipelines using Llama-3.2-1B.
EPIC quantifies reward differences without policy optimization.
problem Distinguishing reward function quality from policy optimization issues.
method EPIC distance to compare reward functions directly.
result EPIC bounds policy training success and regret.
Reinforcement Learning methods are capable of solving complex problems, but resulting policies might perform poorly in environments that are even slightly different. In robotics especially, training and deployment conditions often vary and data collection is expensive, making retraining undesirable. Simulation training…
ICIL learns policies invariant to multiple environments, improving generalization.
problem Learning policies from multiple environments leads to spurious correlations.
method ICIL learns invariant feature representations and a matching imitation policy.
result ICIL policies generalize better to unseen environments.
DR-RPO optimizes robust policies in RL with limited interaction, achieving sublinear regret.
problem Policy optimization in RL under distribution shift and adversarial dynamics.
method DR-RPO algorithm incorporating reference-policy regularization and upper confidence bonus for exploration.
result DR-RPO achieves sublinear regret and polynomial suboptimality bounds in robust RL.
Deep RL trains a robust humanoid push-recovery policy.
problem Training robust humanoid push-recovery policies.
method Model-free Deep Reinforcement Learning.
result Policy learns robust behaviors across the entire body.
iDAD uses neural networks to quickly adapt experiments without likelihoods.
problem Performing adaptive experiments in real-time with implicit models.
method iDAD learns a design policy network upfront to make quick design decisions.
result iDAD can make design decisions in milliseconds, unlike traditional BOED approaches.
AR design reduces multi-armed bandit experiment cost.
problem Comparing multi-armed bandit algorithms requires many restarts.
method Artificial Replay (AR) design for bandit experiments.
result AR reduces experimental cost by requiring only T+o(T) user interactions. Paper uses DRL to optimize portfolios, balancing risk and return.
problem Optimizing portfolios under market uncertainty and risk constraints.
method Integrates Sharpe ratio-based reward with risk control mechanisms, uses PPO for adaptive asset allocation.
result DRL agent stabilizes volatility but sacrifices risk-adjusted returns.
Improved RL policies from offline data with relaxed BC constraints.
problem Overestimation bias in offline RL due to lack of interaction with environment.
method Introducing a policy constraint via behavioural cloning (BC) and adjusting the balance between RL and BC.
result Refined policies outperform baseline and match/exceed complex alternatives.
New RL algorithm tackles online robust MDPs with uncertainty.
problem Developing robust reinforcement learning models for real-world environments.
method Proposes a robust optimistic policy optimization algorithm for online robust MDPs.
result Establishes the first regret bound for online robust MDPs.
Study one-shot strategic classification under unknown costs, improving worst-case accuracy.
problem Learning robust decision rules in strategic settings with unknown user costs.
method Formal study of one-shot strategic classification, framing as a minimax problem, designing efficient algorithms for full-batch and stochastic settings.
result Proves efficient algorithms converge to minimax solution, revealing dual norm regularization's value.
NM-PPG optimizes adaptive feature acquisition in POMDPs for better predictions.
problem Optimizing adaptive feature acquisition in prediction problems with costly features.
method Non-myopic pathwise policy gradients (NM-PPG) with continuous relaxation and straight-through rollout.
result NM-PPG outperforms state-of-the-art AFA methods on synthetic and real-world datasets.
New approach uses SPG for semantic communication without a known channel model.
problem Designing efficient semantic communication systems without a known channel model.
method Applying Stochastic Policy Gradient (SPG) for reinforcement learning.
result Achieves comparable performance to model-aware approaches with a decreased convergence rate.
Deployment-complete benchmarking assesses if evidence leads to consistent deployment actions.
problem Lack of clear evidence leading to consistent deployment actions.
method Introduces deployment-complete benchmarking to test if benchmark evidence determines deployment actions.
result Benchmark evidence must be complete for a claim to lead to a consistent deployment action.
MA-COPP predicts multi-agent system outcomes using data from a different policy, with probabilistic guarantees.
problem Predicting outcomes in multi-agent systems using data from a different policy.
method Conformal prediction framework applied to multi-agent systems, avoiding exhaustive search.
result Achieves probabilistic guarantees for multi-agent system predictions.