April 2026

Reinforcement learning (RL) offers a compelling data-driven paradigm for synthesizing controllers for complex systems when accurate physical models are unavailable; however, most existing control-oriented RL methods assume stationarity and, therefore, struggle in real-world non-stationary deployments where system dynamics and operating conditions can change unexpectedly. Moreover, RL controllers acting in physical environments must satisfy safety constraints throughout their learning and execution phases, rendering transient violations during adaptation unacceptable. Although continual RL and safe RL have each addressed non-stationarity and […]

Ver mais

Personalized Federated Learning (PFL) aims to learn multiple task-specific models rather than a single global model across heterogeneous data distributions. Existing PFL approaches typically rely on iterative optimization-such as model update trajectories-to cluster users that need to accomplish the same tasks together. However, these learning-dynamics-based methods are inherently vulnerable to low-quality data and noisy labels, as corrupted updates distort clustering decisions and degrade personalization performance. To tackle this, we propose FB-NLL, a feature-centric framework that decouples user clustering […]

Ver mais

Explainable artificial intelligence (XAI) has predominantly focused on generating model-centric explanations that approximate the behavior of black-box models. However, such explanations often overlook a fundamental aspect of interpretability: different users require different explanations depending on their goals, preferences, and cognitive constraints. Although recent work has explored user-centric and personalized explanations, most existing approaches rely on heuristic adaptations or implicit user modeling, lacking a principled framework for representing and learning individual preferences. In this paper, we consider Preference-Based Explainable […]

Ver mais

Reinforcement learning-based control policies have been frequently demonstrated to be more effective than analytical techniques for many manipulation tasks. Commonly, these methods learn neural control policies that predict end-effector pose changes directly from observed state information. For tasks like inserting delicate connectors which induce force constraints, pose-based policies have limited explicit control over force and rely on carefully tuned low-level controllers to avoid executing damaging actions. In this work, we present hybrid position-force control policies that learn to […]

Ver mais

Damage identification is a core task in structural health monitoring. In practice, however, its reliability is often compromised by confounding non-damage effects, such as variations in excitation and environmental conditions, which can induce changes comparable to or larger than those caused by structural damage. To address this challenge, this study proposes a self-supervised label-free disentangled representation learning framework for robust vibration-based structural damage identification. The proposed framework employs an autoencoder with two latent representations to learn directly from […]

Ver mais

AI agents are already too human. Not in the romantic sense, not because they love or fear or dream, but in the more banal and frustrating one. The current implementations keep showing their human origin again and again: lack of stringency, lack of patience, lack of focus. Faced with an awkward task, they drift towards the familiar. Faced with hard constraints, they start negotiating with reality. — Andreas Påhlsson-Notini, Less human AI agents, please. Tags: ai-agents, coding-agents, ai

Ver mais

The problem almost nobody warns you about If you’ve tried training a manipulation policy in Isaac Sim or MuJoCo on assets pulled from Sketchfab, TurboSquid, Objaverse, or your team’s internal CAD library, you’ve probably hit one or more of these: Gripper passes through the object. Object has “infinite mass” and refuses to move. Stacking collapses in bizarre non-physical ways. Contact forces spike to NaN and the sim explodes. Your policy trains to 99% success in sim and faceplants […]

Ver mais