AIF reformulated as convex MDP for adaptive behavior.
problem Adaptive behavior and policy optimization.
method Formulating AIF as convex MDP, deriving mirror descent algorithm.
result EFE minimization in AIF is equivalent to reward maximization in latent MDP, with epistemic component.
New algorithm achieves asymptotically optimal regret without horizon dependence.
problem Horizon-free regret minimization for reinforcement learning.
method Proposes a new algorithm and proves a regret upper bound.
result Regret upper bound of \(\tilde O(\sqrt{SAK} + S^8A^3)\) with failure probability \(\delta\).
Study non-asymptotic BPI guarantees for online RL.
problem Identify optimal policy in MDP with high confidence.
method Non-asymptotic sample complexity guarantees for NaS algorithm.
result Sample complexity depends on MDP connectivity and curvature.
Proposes proactive bed requests to reduce ED boarding and patient wait times.
problem Reduces ED boarding and patient wait times by proactively requesting inpatient beds.
method Formulates as a Markov decision process, uses predictions of admission probability and time to disposition to guide bed requests.
result Proactive aggregate bed requests can reduce boarding times by 30-70% and length of stay by 6-15%.
NeuralChaos efficiently approximates complex stochastic processes.
problem Representing and computing square-integrable predictable processes over time.
method Introduces NeuralChaos, a neural operator architecture for R d \mathbb{R}^{d} R d -valued predictable processes. result NeuralChaos achieves best N N N -term chaoslet approximation rates and is dense in H T 2 ( R d ) \mathcal{H}^2_T(\mathbb{R}^{d}) H T 2 ( R d ) . DAGR improves navigation by refining goal representations conditioned on the current state.
problem Goal-conditioned reinforcement learning lacks state awareness, leading to inefficient policy recovery.
method DAGR refines static goal embeddings into state-conditioned ones using gated cross-attention with a state-goal discrepancy map.
result DAGR improves navigation tasks on OGBench, matching or outperforming base methods.
Develops a new reinforcement learning framework for complex control problems.
problem Continuous-time extended mean field control with deterministic policies.
method Model-free sensitivity formula, deterministic policy gradient, local value and advantage-rate representations.
result Demonstrates efficiency, stability, and robustness in solving complex control problems.
DQN outperforms static policies in a dynamic fee environment for automated market makers.
problem How automated market makers (AMMs) perform under dynamic fees is unknown.
method Constructed a closed-loop simulator with dynamic fees, noise flow, and arbitrage.
result A small DQN policy outperforms static policies in a dynamic fee environment.
This paper improves reinforcement learning by estimating return distributions using quantiles.
problem Improving reinforcement learning by estimating return distributions.
method Quantile-based distributional reinforcement learning, using quantile-projected distributional Bellman equations.
result The quantile-based approach achieves optimal sample efficiency and asymptotic efficiency.
This paper improves reinforcement learning by estimating return distributions using quantiles.
problem Improving reinforcement learning by estimating return distributions.
method The paper uses quantile-based distributional reinforcement learning to characterize return distributions.
result The quantile-based approach achieves optimal sample efficiency and asymptotic efficiency.
This paper identifies drift Lipschitz budget K as key to diffusion policy expressivity and statistical trade-offs.
problem Understanding and maximizing the expressivity of diffusion policies while managing statistical limitations.
method Identifying drift Lipschitz budget K as central, quantifying expressivity and statistical behavior, proving lower bounds, and providing practical implementation guidelines.
result Balancing expressivity and statistical complexity yields a finite-sample performance gap, with rates depending on sample size and drift type.
Survey of mathematical foundations for reinforcement learning.
problem Design and analysis of modern reinforcement learning algorithms.
method Organizes mathematical structures from probability, optimization, and operator theory.
result Unified mathematical entry point for researchers in various fields.
EHR-MPC optimizes sepsis treatment using digital twins and inference-time control.
problem Optimal sepsis treatment policies are contested and difficult to adapt during inference.
method EHR-MPC decouples learning patient dynamics from treatment optimization, enabling inference-time control over learned digital twins.
result EHR-MPC achieves comparable off-policy performance and improved simulation performance compared to RL baselines.
FORE evaluates occupancy ratios without requiring Bellman completeness.
problem Offline reinforcement learning occupancy ratio estimation.
method Fitted occupancy-ratio evaluation (FORE) using adjoint Bellman recursion.
result FORE achieves convergence in KL without Bellman completeness.
TREK uses distillation to help students solve hard problems.
problem Stalled progress on hard prompts when current policy lacks useful reasoning trajectories.
method TREK combines distillation and reinforcement learning to expand student support.
result TREK significantly improves student performance on mathematical reasoning and agentic tasks.
LF-IBIS learns optimal policies online without explicit likelihood.
problem Bayesian RL challenges due to intractable likelihood functions.
method Combines ABC with IBIS for online belief updates.
result Approximates posterior distributions for policies and parameters.
Active-GRPO improves molecular optimization by actively deciding when to imitate or self-improve.
problem Training robust and efficient molecular optimization models with large language models.
method Active-GRPO combines imitation and reinforcement learning, upgrading references and policies dynamically.
result Improves molecular optimization performance, achieving statistically significant gains.
Transforming sparse outcomes into dense process rewards for efficient reinforcement learning.
problem Training RL policies to maximize sparse outcomes.
method Incentivizing policy matching state-action visitations of successful episodes.
result Significantly faster RL finetuning performance.
Adversarial attacks on probabilistic state-space models affect latent state and policy decisions.
problem Robust reinforcement learning under adversarial observability.
method Analyzing adversarial attacks on linear probabilistic state-space models.
result Demonstrating the influence of adversarial observations on latent state and policy decisions.
Quantum Annealing Enhanced Reinforcement Learning for Accurate RUL Prediction
problem RUL estimation in predictive maintenance
method QAQL framework combining quantum annealing and Q-learning
result Outperforms classical and quantum baselines
Variational Proximal Policy Optimization improves reinforcement learning from human feedback.
problem Policy mode collapse and brittle exploration loops in reinforcement learning.
method Particle-based variational inference framework with Mixture-of-Experts architecture.
result Significant improvements in complex reasoning benchmarks.
The paper proposes a principle for dynamically adjusting the granularity of reinforcement learning abstractions.
problem Lack of general principles for dynamically adjusting the granularity of reinforcement learning abstractions.
method The paper proposes a principle based on rate-distortion theory, formalized through a performance certificate decomposing value error into learning and abstraction error bounds.
result Soft state-action abstractions can achieve near-optimal performance under substantial lossy compression of state and action information.
A new RL framework allows removing user data without affecting performance.
problem Efficiently removing user data from a reinforcement learning model without affecting performance.
method Formulated a ρ ρ ρ -TV-stable RL algorithm for tabular MDPs that supports exact unlearning. result Achieved a nearly minimax optimal regret bound of Ω ( H S A T + S A H / ρ ) Ω(H\sqrt{\!SAT}\! +\! {SAH}/ρ) Ω ( H S A T + S A H / ρ ) for ρ ρ ρ -TV-stable RL algorithms. ReSkill reconciles RL skill creation with policy optimization.
problem RL policies lack reusable strategies across tasks.
method Integrates skill creation into RL loop with three mechanisms.
result Consistently outperforms existing methods, especially on unseen tasks.
Transformer learns to search through reinforcement learning, mimicking DFS.
problem Understanding how transformers learn search capabilities in RL.
method Two-head transformer, depth-wise curriculum, discounted returns.
result Transformer policy generalizes depth and prioritizes high-probability branches.
Unified analysis of stochastic iterative algorithms using Lyapunov functions.
problem Analyzing convergence of stochastic iterative algorithms for fixed-point equations.
method Lyapunov-based techniques for finite-time analysis of stochastic approximation algorithms.
result Unified mean-square convergence guarantees for various algorithms.
Linear recurrent networks explain reinforcement learning performance in partially observable settings.
problem Understanding why linear recurrent networks work in reinforcement learning with partial observability.
method Constructed and studied two linear filters for HMMs and action-controlled HMMs.
result Linear filters serve as sufficient statistics and reduce state ambiguity, explaining empirical reinforcement learning success.
This study improves convergence of two-timescale SA under Markovian noise in reinforcement learning.
problem Stability and convergence of two-timescale stochastic approximations under Markovian noise.
method Introduced a new control strategy for the fast timescale parameter.
result Established almost sure convergence of TDC with eligibility traces under off-policy learning with linear function approximation.
Physics-informed GCRL tackles sparse feedback learning with hybrid dynamics.
problem Sparse feedback learning with high-dimensional, hybrid, or contact-dependent dynamics.
method Introduces physics-informed inductive biases into goal-conditioned value learning.
result Contact-rich manipulation tasks degrade existing Pi-GCRL methods.
Improved reasoning model by sampling from power distribution without additional training.
problem Efficiently sampling from a sharpened distribution to improve reasoning models.
method Entropy-Cut Metropolis-Hastings algorithm that identifies key decision points for resampling.
result The method consistently improves reasoning models across various datasets.
New approach reduces simulator exploitation by improving strategic robustness.
problem Simulator exploitation leading to reality gap between simulation and real-world performance.
method Formulated as a zero-sum minimax game, providing theoretical guarantees and a convergent active data selection algorithm.
result Proves convergence and reduces prediction error in strategically important regions by 1.5-2.2 times.
New approach reduces simulator exploitation by learning robust models.
problem Simulator exploitation leading to reality gap in reinforcement learning.
method Formulated as a zero-sum minimax game between model player and policy player, providing theoretical guarantees and a convergent active data selection algorithm.
result Reduces prediction error in strategically important regions by 1.5-2.2 times and enables near-optimal real-world performance.
New algorithm reduces reinforcement learning regret by adapting to interaction variability.
problem Existing reinforcement learning methods lack adaptability to interaction variability.
method Developed a variance-adaptive optimal algorithm for MNL function approximation.
result Achieved instance-wise optimal regret bounds, validating efficiency in practice.
This work shows synthetic gradients can outperform backpropagation in sample efficiency.
problem The efficiency of backpropagation in training neural networks.
method Unified vectorized feedback framework for loss-based and reward-based learning, introducing synthetic gradients.
result Synthetic gradients can achieve lower gradient-estimation mean squared error than backpropagation under certain conditions.
New approach transfers rewards learned in one environment to reinforcement learning in a new environment.
problem Transfer of rewards learned using inverse reinforcement learning from one environment to a new, different environment.
method Formulate the problem as a joint system of Bellman equations, develop minimax estimators for the target soft- q q q -function, solve the source and target system of equations jointly. result The coupled approach removes the first-order influence of source Bellman residual error compared to the sequential approach.
Surrogate models speed up RL training in dynamic systems.
problem High computational cost of high-fidelity simulations.
method Developed and tested surrogate models for RL training.
result Surrogate models can significantly accelerate RL training.
BASIS improves LLM reasoning by sharing batchwise rollout info, reducing MSE by 69%.
problem Improving large language model reasoning with limited rollouts and batch information.
method BASIS samples only one rollout per prompt but uses batch information to improve value function estimation.
result BASIS reduces MSE in value function estimation by 69% compared to REINFORCE++.
A new method solves bilevel optimization problems in competitive Markov games.
problem Capturing competitive structures in RL with multiple interacting policies.
method Penalty-augmented Nikaido-Isoda descent-ascent (PANDA) method.
result PANDA converges to stationary points without convexity assumptions.
A new method reduces variance in training early-stage rankers for large-scale search systems.
problem Training early-stage rankers for large-scale search systems is challenging due to exploding variance in policy gradient methods.
method Proposes credit-assigned policy gradient (CA-PG) to mitigate variance in training early-stage rankers.
result CA-PG significantly reduces variance in training early-stage rankers compared to vanilla policy gradient.
No policy can simultaneously be fully autonomous, optimally calibrated, and helpful, proving a trilemma.
problem Proving impossibility of a policy achieving maximum helpfulness, optimal calibration, and full autonomy.
method Geometric proof showing that adding any non-affine autonomy incentive to a strictly proper scoring rule destroys strict properness.
result The Behavioral Credibility Trilemma: no policy can achieve all three goals simultaneously.
Aims to create safe reinforcement learning policies by considering individual harm.
problem Optimal policies for a population may harm certain individuals.
method Formalizes individual harm, proposes a two-stage procedure, and establishes finite-sample properties.
result Learned policies maximize expected return while minimizing harm.
Study preference-based reinforcement learning in episodic kernel MDPs.
problem Learning from episodic human preferences in reinforcement learning.
method Developed preference-based value estimation and confidence sets for kernel-based MDPs.
result Proved high-probability regret bounds that converge to optimal policy value.
Study risk-sensitive reinforcement learning with optimized certainty equivalents.
problem Risk-sensitive reinforcement learning in finite discounted MDPs.
method Analyzed a simple model-based approach and derived PAC sample complexity bounds.
result Established tight sample complexity bounds for value and policy learning.
ABS dynamically adjusts batch size based on policy stability, improving RL performance.
problem Diminishing returns with large batch sizes in RL due to non-stationary data.
method Adaptive Batch Scaling (ABS) with Behavioral Divergence metric.
result Larger batch sizes can improve RL performance, contrary to conventional wisdom.
Improved regret bound for MNL MDPs with variance-aware approach.
problem Optimal reinforcement learning for MNL MDPs with structured variance.
method Introducing a problem-dependent constant measuring average variance, proposing an algorithm with improved regret bound.
result Minimax optimal regret bound of O ( d H 2 σ ˉ T T ) O(dH^2\barσ_T\sqrt{T}) O ( d H 2 σ ˉ T T ) for structured MDPs. Paper uses RL to optimize daily step distribution for better health biomarkers.
problem Lack of personalized PA distribution recommendations for health biomarkers.
method Developed an offline reinforcement learning algorithm to learn optimal PA distributions.
result Learned optimal policy suggests more consistent daily steps and tailored recommendations.
NeuroMAS treats multi-agent systems as neural networks for scalable, trainable coordination.
problem Designing multi-agent systems as hand-designed workflows is inefficient and inflexible.
method NeuroMAS treats multi-agent systems as a neural network architecture with reinforcement learning for scalable coordination.
result NeuroMAS improves significantly over multi-agent baselines and can be scaled progressively.
Improved reinforcement learning for episodes with varying action sets.
problem Reinforcement learning with context-dependent action sets.
method Extends MVP algorithm to handle adversarial and stochastic contexts.
result Established minimax regret bounds of O ( S A H 3 K log L ) O(\sqrt{SAH^3K\log L}) O ( S A H 3 K log L ) for adversarial contexts. Hybrid Policy Optimization tackles reinforcement learning in hybrid spaces, improving performance over PPO.
problem Credit assignment issues and biased gradients in hybrid discrete-continuous action spaces.
method Mixed gradient estimator combining pathwise and score-function gradients, reformulating problems in hybrid form.
result HPO substantially outperforms PPO on inventory control and switched systems, with performance gaps increasing with continuous action dimension.
Improved sample complexity for identifying best policies in risk-sensitive reinforcement learning.
problem Identifying approximately optimal policies in risk-sensitive reinforcement learning with exponential horizon dependence.
method Forward-model based algorithm with KL-based exploration bonuses adapted for entropic criterion, leveraging smoothness properties of exponential utility and a new stopping rule.
result Achieved sample complexity matching the lower bound, closing the gap between upper and lower bounds.
Paper achieves ε − 2 ε^{-2} ε − 2 sample complexity for actor-critic methods with minimal assumptions.
problem Achieving ε − 2 ε^{-2} ε − 2 sample complexity for actor-critic methods under minimal assumptions. method Single-loop, single-timescale implementation; coupled Lyapunov drift framework.
result First i l d e O ( ε − 2 ) ilde{\mathcal{O}}(ε^{-2}) i l d e O ( ε − 2 ) sample complexity guarantee for finding an ε ε ε -optimal policy. Enhances reinforcement learning from sparse data.
problem Limited data for offline reinforcement learning.
method Trajectory-based data augmentation.
result Improves reinforcement learning performance.
AIS corrects rollout-training mismatch in quantized RL, improving speed and stability.
problem Rollout-training mismatch in quantized RL causes bias and training collapse.
method Adaptive Importance Sampling (AIS) adjusts gradient correction per batch.
result AIS matches BF16 baseline on most tasks while improving speed.
Optimizes trading policies using future price forecasts.
problem Static reinforcement learning agents lack mechanisms for using price forecasts at inference time.
method FPILOT framework inspired by Model Predictive Control (MPC). Uses a predictive model to construct an allocation-based imagined return objective at each decision step.
result Consistent improvements in total return and risk-adjusted metrics across various policy learning algorithms.
Bootstrap method for Markov chains in reinforcement learning.
problem Distributional consistency in finite controlled Markov chains with unknown control policies.
method Model-based bootstrap with novel LLN and CLT for visitation counts and transition increments.
result Asymptotically valid confidence intervals for value and Q Q Q -functions in offline RL. A new method resolves non-identifiability in reward modeling using anchor labels.
problem Non-identifiability in reward modeling from pairwise preferences alone.
method Anchor-guided Variance-aware Reward Modeling (AVRM) framework.
result AVRM resolves non-identifiability and improves reward modeling performance.
TOPPO improves PPO for MTRL by balancing critic gradients, outperforming SAC.
problem Critic-side gradient ill-conditioning in PPO for MTRL.
method Critic Balancing modules to improve gradient conditioning and balance task updates.
result TOPPO achieves stronger mean and tail-task performance than SAC-family and ARS-family baselines.
DSPI connects natural policy gradient to policy iteration, proving global convergence.
problem Optimizing policies in reinforcement learning.
method DSPI framework, combining smoothed policy iteration and natural policy gradient.
result DSPI achieves geometric convergence and optimal complexity for policy optimization.
Unified framework for adaptive learning systems using consolidation and expansion operations.
problem Managing the balance between consolidating known knowledge and expanding into new evidence in adaptive learning systems.
method Introduces Consolidation-Expansion Operator Mechanics (OpMech) with the order-gap metric to control the balance.
result The order-gap signal provides real-time control and termination guarantees for adaptive learning systems.
RLMM extends psychometric models to larger tasks.
problem Sequential process data from interactive assessments are not well handled by conventional models.
method RLMM decouples person-level choice sensitivity from task-level value representation through a shared parametric action-value function.
result RLMM achieves higher estimation accuracy and lower runtime than MDP-MM in peg-solitaire simulations and AQUALAB gameplay logs.