Traditional reinforcement learning (RL) assumes the agents make decisions based on Markov decision processes (MDPs) with one-step transition models. In many real-world applications, such as energy management and stock investment, agents can access multi-step predictions of future states, which provide additional advantages for decision making. However, multi-step predictions are inherently high-dimensional: naively embedding these predictions into an MDP leads to an exponential blow-up in state space and the curse of dimensionality. Moreover, existing RL theory provides few tools to analyze prediction-augmented MDPs, as it typically works on one-step transition kernels and cannot accommodate multi-step predictions with errors or partial action-coverage. We address these challenges with three key innovations: First, we propose the \emph{Bayesian value function} to characterize the optimal prediction-aware policy tractably. Second, we develop a novel \emph{Bellman-Jensen Gap} analysis on the Bayesian value function, which enables characterizing the value of imperfect predictions. Third, we introduce BOLA (Bayesian Offline Learning with Online Adaptation), a two-stage model-based RL algorithm that separates offline Bayesian value learning from lightweight online adaptation to real-time predictions. We prove that BOLA remains sample-efficient even under imperfect predictions. We validate our theory and algorithm on synthetic MDPs and a real-world wind energy storage control problem.

We present an online model-based reinforcement learning algorithm suitable for controlling complex robotic systems directly in the real world. Unlike prevailing sim-to-real pipelines that rely on extensive offline simulation and model-free policy optimization, our method builds a dynamics model from real-time interaction data and performs policy updates guided by the learned dynamics model. This efficient model-based reinforcement learning scheme significantly reduces the number of samples to train control policies, enabling direct training on real-world rollout data. This significantly reduces the influence of bias in the simulated data, and facilitates the search for high-performance control policies. We adopt online learning analysis to derive sublinear regret bounds under standard stochastic online optimization assumptions, providing formal guarantees on performance improvement as more interaction data are collected. Experimental evaluations were performed on a hydraulic excavator arm and a soft robot arm, where the algorithm demonstrates strong sample efficiency compared to model-free reinforcement learning methods, reaching comparable performance within hours. Robust adaptation to shifting dynamics was also observed when the payload condition was randomized. Our approach paves the way toward efficient and reliable on-robot learning for a broad class of challenging control tasks.

Reliable navigation in safety-critical environments requires both accurate hazard perception and principled uncertainty handling to strengthen downstream safety handling. Despite the effectiveness of existing approaches, they assume perfect hazard detection capabilities, while uncertainty-aware perception approaches lack finite-sample guarantees. We present COPPOL, a conformal-driven perception-to-policy learning approach that integrates distribution-free, finite-sample safety guarantees into semantic segmentation, yielding calibrated hazard maps with rigorous bounds for missed detections. These maps induce risk-aware cost fields for downstream RL planning. Across two satellite-derived benchmarks, COPPOL increases hazard coverage (up to 6x) compared to comparative baselines, achieving near-complete detection of unsafe regions while reducing hazardous violations during navigation (up to approx 50%). More importantly, our approach remains robust to distributional shift, preserving both safety and efficiency.

Ensuring the safe exploration of reinforcement learning (RL) agents is critical for deployment in real-world systems. Yet existing approaches struggle to strike the right balance: methods that tightly enforce safety often cripple task performance, while those that prioritize reward leave safety constraints frequently violated, producing diffuse cost landscapes that flatten gradients and stall policy improvement. We introduce the Uncertain Safety Critic (USC), a novel approach that integrates uncertainty-aware modulation and refinement into critic training. By concentrating conservatism in uncertain and costly regions while preserving sharp gradients in safe areas, USC enables policies to achieve effective reward-safety trade-offs. Extensive experiments show that USC reduces safety violations by approximately 40% while maintaining competitive or higher rewards, and reduces the error between predicted and true cost gradients by approximately 83%, breaking the prevailing trade-off between safety and performance and paving the way for scalable safe RL.

Short video streaming has become a dominant paradigm in digital media, characterized by rapid swiping interactions and diverse media content. A key technical challenge is designing an effective preloading strategy that dynamically selects and prioritizes download tasks from an evolving playlist, balancing Quality of Experience (QoE) and bandwidth efficiency under practical commercial constraints. However, real world analysis reveals critical limitations of existing approaches: (1) insufficient adaptation of download task sizes to dynamic conditions, and (2) watch time prediction models that are difficult to deploy reliably at scale. In this paper, we propose DeLoad, a novel preloading framework that addresses these issues by introducing dynamic task sizing and a practical, multi dimensional watch time estimation method. Additionally, a Deep Reinforcement Learning (DRL) enhanced agent is trained to optimize the download range decisions adaptively. Extensive evaluations conducted on an offline testing platform, leveraging massive real world network data, demonstrate that DeLoad achieves significant improvements in QoE metrics (34.4% to 87.4% gain). Furthermore, after deployment on a large scale commercial short video platform, DeLoad has increased overall user watch time by 0.09% while simultaneously reducing rebuffering events and 3.76% bandwidth consumption.

Self-play constitutes a fundamental paradigm for autonomous skill acquisition, whereby agents iteratively enhance their capabilities through self-directed environmental exploration. Conventional self-play frameworks exploit agent symmetry within zero-sum competitive settings, yet this approach proves inadequate for open-ended learning scenarios characterized by inherent asymmetry. Human pedagogical systems exemplify asymmetric instructional frameworks wherein educators systematically construct challenges calibrated to individual learners' developmental trajectories. The principal challenge resides in operationalizing these asymmetric, adaptive pedagogical mechanisms within artificial systems capable of autonomously synthesizing appropriate curricula without predetermined task hierarchies. Here we present Heterogeneous Adversarial Play (HAP), an adversarial Automatic Curriculum Learning framework that formalizes teacher-student interactions as a minimax optimization wherein task-generating instructor and problem-solving learner co-evolve through adversarial dynamics. In contrast to prevailing ACL methodologies that employ static curricula or unidirectional task selection mechanisms, HAP establishes a bidirectional feedback system wherein instructors continuously recalibrate task complexity in response to real-time learner performance metrics. Experimental validation across multi-task learning domains demonstrates that our framework achieves performance parity with SOTA baselines while generating curricula that enhance learning efficacy in both artificial agents and human subjects.

The classical policy gradient method is the theoretical and conceptual foundation of modern policy-based reinforcement learning (RL) algorithms. Most rigorous analyses of such methods, particularly those establishing convergence guarantees, assume a discount factor $γ< 1$. In contrast, however, a recent line of work on policy-based RL for large language models uses the undiscounted total-reward setting with $γ= 1$, rendering much of the existing theory inapplicable. In this paper, we provide analyses of the policy gradient method for undiscounted expected total-reward infinite-horizon MDPs based on two key insights: (i) the classification of the MDP states into recurrent and transient states is invariant over the set of policies that assign strictly positive probability to every action (as is typical in deep RL models employing a softmax output layer) and (ii) the classical state visitation measure (which may be ill-defined when $γ= 1$) can be replaced with a new object that we call the transient visitation measure.

Finding Nash equilibria in imperfect-information games remains a central challenge in multi-agent reinforcement learning. While regularization-based methods have recently achieved last-iteration convergence to a regularized equilibrium, they require the regularization strength to shrink toward zero to approximate a Nash equilibrium, often leading to unstable learning in practice. Instead, we fix the regularization strength at a large value for robustness and achieve convergence by iteratively refining the reference policy. Our main theoretical result shows that this procedure guarantees strictly monotonic improvement and convergence to an exact Nash equilibrium in two-player zero-sum games, without requiring a uniqueness assumption. Building on this framework, we develop a practical algorithm, Nash Policy Gradient (NashPG), which preserves the generalizability of policy gradient methods while relying solely on the current and reference policies. Empirically, NashPG achieves comparable or lower exploitability than prior model-free methods on classic benchmark games and scales to large domains such as Battleship and No-Limit Texas Hold'em, where NashPG consistently attains higher Elo ratings.

Recent advances in reinforcement learning (RL) enable its use on increasingly complex tasks, but the lack of formal safety guarantees still limits its application in safety-critical settings. A common practical approach is to augment the RL policy with a safety filter that overrides unsafe actions to prevent failures during both training and deployment. However, safety filtering is often perceived as sacrificing performance and hindering the learning process. We show that this perceived safety-performance tradeoff is not inherent and prove, for the first time, that enforcing safety with a sufficiently permissive safety filter does not degrade asymptotic performance. We formalize RL safety with a safety-critical Markov decision process (SC-MDP), which requires categorical, rather than high-probability, avoidance of catastrophic failure states. Additionally, we define an associated filtered MDP in which all actions result in safe effects, thanks to a safety filter that is considered to be a part of the environment. Our main theorem establishes that (i) learning in the filtered MDP is safe categorically, (ii) standard RL convergence carries over to the filtered MDP, and (iii) any policy that is optimal in the filtered MDP-when executed through the same filter-achieves the same asymptotic return as the best safe policy in the SC-MDP, yielding a complete separation between safety enforcement and performance optimization. We validate the theory on Safety Gymnasium with representative tasks and constraints, observing zero violations during training and final performance matching or exceeding unfiltered baselines. Together, these results shed light on a long-standing question in safety-filtered learning and provide a simple, principled recipe for safe RL: train and deploy RL policies with the most permissive safety filter that is available.

W e present UniRL-Zero, a unified reinforcement learning (RL) framework that boosts, multimodal language model understanding and reasoning, diffusion model multimedia generation, and their beneficial interaction capabilities within a unified model. Our work defines six scenarios for unified model reinforcement learning, providing systematic baselines for reinforcement learning of unified understanding and generation model. Our code is available at https://github.com/G-UN/UniRL.