reinforcement learning

RL without Tears: An Introduction in the Era of LLMs

Reinforcement learning (RL) has become a powerful and versatile paradigm in the era of large language models (LLMs). It is widely used to align models with human preferences and improve their reasoning capabilities. More recently, with the rapid rise of LLM-based agents, RL has become increasingly important for enablin...

RAVEN: Receiver-conditioned Action-Value ENcoding for Finite-Symbol Multi-Agent Communication

Reference implementation of RAVEN, a finite-symbol (K = 4, 2 bits per message) communication method for cooperative multi-agent reinforcement learning. A sender's symbol is chosen so that, in the receiver's own decision context, the receiver's centred action-value profile matches the one it would have with full informa...

Rater State Bias in RLHF Preference Data: An Audit Framework

We identify a structured confound in Reinforcement Learning from Human Feedback (RLHF). Pairwise preference labels are intended to reflect the compared outputs. They may also reflect the rater's state during annotation. Under sustained stressful or distressing conditions, a rater's preferences may shift over time, so t...

Knowledge-Aware Technician Allocation: The Long-Term Impact of Technician Upskilling (code)

Technician allocation in industrial maintenance is a domain-specific Human Resource Allocation Problem (HRAP). Yet most existing approaches treat worker skills as static, overlooking how skill development and fatigue shape long-term operational performance. We extend HRAP to a sequential Reinforcement Learning (RL) set...

From awareness to focus: using digital mindfulness and learning strategies to combat cognitive overload in the hybrid knowledge economy

Purpose This paper discusses digital mindfulness, a significant intervention to reduce the attention fatigue and cognitive overload in hybrid knowledge economy. Design/methodology/approach The study uses a conceptual approach focusing on mindfulness theory and the stages of change model to design and implement the lear...

Postmodern Physics of Hamzah Information.(308)

استدلال ۹۱ از ۱۰۰: ابطال منطقی الگوریتم‌های یادگیری تقویتی و فروپاشی توابع پاداش احتمالی ($\text{The Logical Invalidation of Reinforcement Learning Algorithms and Reward Collapse}$) ۱. مقدمه و طرح صورت تفصیلی معما ($\text{Reinforcement Learning Invalidation Paradox Setup}$) در تمامی چارچوب‌های یادگیری ماشین مرسوم، یادگ...

Dataset for Reinforcement learning enhanced by fractional differencing for portfolio optimization

GCN-based learner movement feature network modeling for personalized training path planning and dynamic evaluation

Abstract Personalized training path planning has emerged as an active concern in intelligent sports education, yet many existing approaches struggle to capture the dependencies that link diverse movement features. This study develops a graph convolutional neural network (GCN) approach that models learner movement featu...

Editorial: Impact of digital addiction on mental health and comorbidities in older adults

Digital technologies have become an integral part of everyday life across the lifespan. Yet research on problematic internet and smartphone use has traditionally concentrated on adolescents and young adults, leaving substantially less evidence on later life. This imbalance has become increasingly difficult to justify....

Hybrid Convolutional Neural Network–Bidirectional Gated Recurrent Unit with Dual Attention and Reinforcement Learning for Rolling Element Bearing Fault Diagnosis

The diagnosis of rolling element bearing faults is of most importance for the reliable and safe operation of rotating machinery. An adaptive selection of fault features based on a hybrid convolutional neural network (CNN)-bidirectional gated recurrent unit (BiGRU) framework with dual attention and reinforcement learnin...

Heat Pump Control in Urban Buildings: Methods and Deployment Barriers

Purpose Heat pumps are central to the decarbonisation of urban building stock, yet their real-world performance in European cities often falls short of nominal efficiency. A major limitation is often not the heat-pump unit itself, but system-level control and integration, including supervisory control, hydraulic integr...

Scheduling and energy management in a renewable-energy-driven microgrid with mobile charging robots: A structure-aware reinforcement learning approach