reinforcement learning

Human outcome-irrelevant value learning is insensitive to its monetary cost

When learning from reinforcement, individuals should assign credit selectively to features of their actions that causally determine outcomes. Yet, even when explicit instructions and extensive experience clarify which action features predict reward, humans assign credit to outcome-irrelevant features, leading to subopt...

Reinforcement learning with experience sharing promotes collective cooperation on higher-order networks

Beyond Traditional Risk Scores: Artificial Intelligence in Coronary Plaque Characterization and Personalized Atherosclerosis Management

This review critically examines the transformative role of artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL) and deep learning (DL), in redefining atherosclerosis management across three interconnected domains and explores how AI-driven predictive models integrate high-dimensional...

Branch-level evaluation of a deep reinforcement learning controller for joint handover, power and spectrum control

Pre-registered, branch-level evaluation of a distributional Dueling N-step DQN with a branching action head for joint handover, transmit power and sub-band control in multi-UE 140 GHz vehicular O-RAN. Includes every pre-registration, including those whose predictions failed.

Deep Reinforcement Learning based Optimal Routing with Software-Defined Networking

An effective method for network congestion avoidance is to optimize the routing. We propose an optimalrouting controlling that combines Deep Reinforcement learning(DRL) based policy optimization andSoftware Defined Network(SDN) based flow control. It uses DRL to learn the optimal network routingpolicy in different traf...

Decodable, Compositional Affect-Like Representations in a Self-Taught Agent and Pretrained Language Models

We study whether emotion-like internal variables arise without hand-coding in artificial networks, and whether affect-related representations share common geometric structure across systems. In a self-taught homeostatic reinforcement-learning agent - trained only with a well-being reward and no emotion labels or object...

PP-FEO: Privacy-Preserving Federated Ensemble Optimization for Intrusion Detection

This repository contains the source code and reproducibility materials associated with the Privacy-Preserving Federated Ensemble Optimization (PP-FEO) framework for intrusion detection in distributed Cyber-Physical System (CPS) environments. The implementation includes data preprocessing and feature harmonization, the...

Beyond Automation: Reimagining Human-Centered Systems for AI-Mediated Institutions

ABSTRACT PurposeThis paper examines how AI and automation can reinforce existing institutional weaknesses when technological efficiency becomes disconnected from human realities, contextual judgment, ethical responsibility, and institutional coherence. It argues that the central challenge is not automation itself, but...

The Adaptation Dilemma: Cultural Fit Does Not Guarantee Safety in Mental-Health LLMs

People increasingly take mental-health concerns to general-purpose AI chatbots. A first wave of safety research has begun to map LLM failure points in this context. When that work considers culture, it usually reads the problem as a lack of fit: if the model does not match the user, the fix is to make it match more clo...

Dissecting aminergic contributions to visual discrimination learning in honey bees using a virtual reality paradigm

The honey bee is a powerful model for studying visual cognition, yet the neuromodulatory mechanisms underlying visual learning remain poorly understood due to the reliance on free-flying paradigms that preclude neural interventions. Here, we used a closed-loop virtual reality (VR) system in which tethered bees walk on...

REMP: A transformer with role-specific experts and multi-scale positional encoding for autonomous multi-UAV air combat

REMP is proposed, a Transformer-based framework that integrates Role-Specific Experts with Multi-Scale Positional Encoding and uses Fourier feature mapping to alleviate spectral bias and capture fine-grained 3D spatial features that are critical for tactical maneuvering.

Optimization of Precision Fermentation Using AI/ML

Precision fermentation represents a frontier convergence of synthetic biology and process engineering, employing genetically modified microbial systems to biosynthesize tailored biomolecules—ranging from proteins, amino acids, and fatty acids to pigments, vitamins, and therapeutic intermediates. The integration of arti...