Group-fairness metrics (e.g., equalized odds) can vary sharply across resamples and are especially brittle under distribution shift, undermining reliable audits. We propose a Wasserstein distributionally robust framework that certifies worst-case group fairness over a ball of plausible test distributions centered at the empirical law. Our formulation unifies common group fairness notions via a generic conditional-probability functional and defines $\varepsilon$-Wasserstein Distributional Fairness ($\varepsilon$-WDF) as the audit target. Leveraging strong duality, we derive tractable reformulations and an efficient estimator (DRUNE) for $\varepsilon$-WDF. We prove feasibility and consistency and establish finite-sample certification guarantees for auditing fairness, along with quantitative bounds under smoothness and margin conditions. Across standard benchmarks and classifiers, $\varepsilon$-WDF delivers stable fairness assessments under distribution shift, providing a principled basis for auditing and certifying group fairness beyond observational data.

We study the problem of generalizing an expert agent's behavior, provided through demonstrations, to new environments and/or additional constraints. Inverse Reinforcement Learning (IRL) offers a promising solution by seeking to recover the expert's underlying reward function, which, if used for planning in the new settings, would reproduce the desired behavior. However, IRL is inherently ill-posed: multiple reward functions, forming the so-called feasible set, can explain the same observed behavior. Since these rewards may induce different policies in the new setting, in the absence of additional information, a decision criterion is needed to select which policy to deploy. In this paper, we propose a novel, principled criterion that selects the "average" policy among those induced by the rewards in a certain bounded subset of the feasible set. Remarkably, we show that this policy can be obtained by planning with the reward centroid of that subset, for which we derive a closed-form expression. We then present a provably efficient algorithm for estimating this centroid using an offline dataset of expert demonstrations only. Finally, we conduct numerical simulations that illustrate the relationship between the expert's behavior and the behavior produced by our method.

Empirical risk minimization (ERM) is a cornerstone of modern machine learning (ML), supported by advances in optimization theory that ensure efficient solutions with provable algorithmic convergence rates, which measure the speed at which optimization algorithms approach a solution, and statistical learning rates, which characterize how well the solution generalizes to unseen data. Privacy, memory, computational, and communications constraints increasingly necessitate data collection, processing, and storage across network-connected devices. In many applications, these networks operate in decentralized settings where a central server cannot be assumed, requiring decentralized ML algorithms that are both efficient and resilient. Decentralized learning, however, faces significant challenges, including an increased attack surface for adversarial interference during decentralized learning processes. This paper focuses on the man-in-the-middle (MITM) attack, which can cause models to deviate significantly from their intended ERM solutions. To address this challenge, we propose RESIST (Resilient dEcentralized learning using conSensus gradIent deScenT), an optimization algorithm designed to be robust against adversarially compromised communication links. RESIST achieves algorithmic and statistical convergence for strongly convex, Polyak-Lojasiewicz, and nonconvex ERM problems. Experimental results demonstrate the robustness and scalability of RESIST for real-world decentralized learning in adversarial environments.

We address the problem of best policy identification in preference-based reinforcement learning (PbRL), where learning occurs from noisy binary preferences over trajectory pairs rather than explicit numerical rewards. This approach is useful for post-training optimization of generative AI models during multi-turn user interactions, where preference feedback is more robust than handcrafted reward models. In this setting, learning is driven by both an offline preference dataset -- collected from a rater of unknown 'competence' -- and online data collected with pure exploration. Since offline datasets may exhibit out-of-distribution (OOD) biases, principled online data collection is necessary. To address this, we propose Posterior Sampling for Preference Learning ($\mathsf{PSPL}$), a novel algorithm inspired by Top-Two Thompson Sampling, that maintains independent posteriors over the true reward model and transition dynamics. We provide the first theoretical guarantees for PbRL in this setting, establishing an upper bound on the simple Bayesian regret of $\mathsf{PSPL}$. Since the exact algorithm can be computationally impractical, we also provide an approximate version that outperforms existing baselines.

Millimeter wave (mmWave)-based orthogonal frequency-division multiplexing (OFDM) stands out as a suitable alternative for high-resolution sensing and high-speed data transmission. To meet communication and sensing requirements, many works propose a static configuration where the wave's hyperparameters such as the number of symbols in a frame and the number of frames in a communication slot are already predefined. However, two facts oblige us to redefine the problem, (1) the environment is often dynamic and uncertain, and (2) mmWave is severely impacted by wireless environments. A striking example where this challenge is very prominent is autonomous vehicle (AV). Such a system leverages integrated sensing and communication (ISAC) using mmWave to manage data transmission and the dynamism of the environment. In this work, we consider an autonomous vehicle network where an AV utilizes its queue state information (QSI) and channel state information (CSI) in conjunction with reinforcement learning techniques to manage communication and sensing. This enables the AV to achieve two primary objectives: establishing a stable communication link with other AVs and accurately estimating the velocities of surrounding objects with high resolution. The communication performance is therefore evaluated based on the queue state, the effective data rate, and the discarded packets rate. In contrast, the effectiveness of the sensing is assessed using the velocity resolution. In addition, we exploit adaptive OFDM techniques for dynamic modulation, and we suggest a reward function that leverages the age of updates to handle the communication buffer and improve sensing. The system is validated using advantage actor-critic (A2C) and proximal policy optimization (PPO). Furthermore, we compare our solution with the existing design and demonstrate its superior performance by computer simulations.

Designing control policies whose performance level is guaranteed to remain above a given threshold in a span of environments is a critical feature for the adoption of reinforcement learning (RL) in real-world applications. The search for such robust policies is a notoriously difficult problem, related to the so-called dynamic model of transition function uncertainty, where the environment dynamics are allowed to change at each time step. But in practical cases, one is rather interested in robustness to a span of static transition models throughout interaction episodes. The static model is known to be harder to solve than the dynamic one, and seminal algorithms, such as robust value iteration, as well as most recent works on deep robust RL, build upon the dynamic model. In this work, we propose to revisit the static model. We suggest an analysis of why solving the static model under some mild hypotheses is a reasonable endeavor, based on an equivalence with the dynamic model, and formalize the general intuition that robust MDPs can be solved by tackling a series of static problems. We introduce a generic meta-algorithm called IWOCS, which incrementally identifies worst-case transition models so as to guide the search for a robust policy. Discussion on IWOCS sheds light on new ways to decouple policy optimization and adversarial transition functions and opens new perspectives for analysis. We derive a deep RL version of IWOCS and demonstrate it is competitive with state-of-the-art algorithms on classical benchmarks.

We consider Thompson Sampling (TS) for linear combinatorial semi-bandits and subgaussian rewards. We propose the first known TS whose finite-time regret does not scale exponentially with the dimension of the problem. We further show the "mismatched sampling paradox": A learner who knows the rewards distributions and samples from the correct posterior distribution can perform exponentially worse than a learner who does not know the rewards and simply samples from a well-chosen Gaussian posterior.

Safety is one of the key issues preventing the deployment of reinforcement learning techniques in real-world robots. While most approaches in the Safe Reinforcement Learning area do not require prior knowledge of constraints and robot kinematics and rely solely on data, it is often difficult to deploy them in complex real-world settings. Instead, model-based approaches that incorporate prior knowledge of the constraints and dynamics into the learning framework have proven capable of deploying the learning algorithm directly on the real robot. Unfortunately, while an approximated model of the robot dynamics is often available, the safety constraints are task-specific and hard to obtain: they may be too complicated to encode analytically, too expensive to compute, or it may be difficult to envision a priori the long-term safety requirements. In this paper, we bridge this gap by extending the safe exploration method, ATACOM, with learnable constraints, with a particular focus on ensuring long-term safety and handling of uncertainty. Our approach is competitive or superior to state-of-the-art methods in final performance while maintaining safer behavior during training.

Model-based Reinforcement Learning (RL) has shown its high training efficiency and capability of handling high-dimensional tasks. Regarding safety issues, safe model-based RL can achieve nearly zero-cost performance and effectively manage the trade-off between performance and safety. Nevertheless, prior works still pose safety challenges due to the online exploration in real-world deployment. To address this, some offline RL methods have emerged as solutions, which learn from a static dataset in a safe way by avoiding interactions with the environment. In this paper, we aim to further enhance safety during the deployment stage for vision-based robotic tasks by fine-tuning an offline-trained policy. We incorporate in-sample optimization, model-based policy expansion, and reachability guidance to construct a safe offline-to-online framework. Moreover, our method proves to improve the generalization of offline policy in unseen safety-constrained scenarios. Finally, the efficiency of our method is validated on simulation benchmarks with five vision-only tasks and a real robot by solving some deployment problems using limited data.

We study the constant regret guarantees in reinforcement learning (RL). Our objective is to design an algorithm that incurs only finite regret over infinite episodes with high probability. We introduce an algorithm, Cert-LSVI-UCB, for misspecified linear Markov decision processes (MDPs) where both the transition kernel and the reward function can be approximated by some linear function up to misspecification level $\zeta$. At the core of Cert-LSVI-UCB is an innovative certified estimator, which facilitates a fine-grained concentration analysis for multi-phase value-targeted regression, enabling us to establish an instance-dependent regret bound that is constant w.r.t. the number of episodes. Specifically, we demonstrate that for an MDP characterized by a minimal suboptimality gap $\Delta$, Cert-LSVI-UCB has a cumulative regret of $\tilde{\mathcal{O}}(d^3H^5/\Delta)$ with high probability, provided that the misspecification level $\zeta$ is below $\tilde{\mathcal{O}}(\Delta / (\sqrt{d}H^2))$. Remarkably, this regret bound remains constant relative to the number of episodes $K$. To the best of our knowledge, Cert-LSVI-UCB is the first algorithm to achieve a constant, instance-dependent, high-probability regret bound in RL with linear function approximation for infinite runs without relying on prior distribution assumptions. This not only highlights the robustness of Cert-LSVI-UCB to model misspecification but also introduces novel algorithmic designs and analytical techniques of independent interest.