reinforcement learning

Weakly Supervised Leaf Disease Segmentation via Reinforcement Learning-Based Prompt Optimization for SAM

Accurate segmentation of leaf lesions is essential for crop disease monitoring and precision agriculture. However, most existing methods rely on pixel-level annotations, which are costly and difficult to obtain in real agricultural scenarios. Weakly supervised approaches alleviate this requirement but often suffer from...

Dynamic attention for container relocation problems: a deep reinforcement learning approach

Experimental Research of Hybrid Reinforcement Learning Methods for QoS-Aware Web Service Composition: an Actor-Critic Approach

A high-order discontinuous Galerkin framework combining volume-penalization immersed boundaries and dynamic p-adaptation for moving-body and fluid-structure interaction simulations

This study presents a high-order numerical framework for fluid-structure interaction (FSI) problems involving rigid motion or deformable elastic solids subjected to low Reynolds number viscous flow. The Navier-Stokes equations in the fluid domain are solved using a high-order nodal discontinuous Galerkin (DG) solver HO...

Global State-Guided Reinforcement Learning-Assisted Differential Evolution for Multi-Objective Optimization

Contact-History-Based Velocity-Tracking Reward Derived from Center-of-Mass Controllability for Quadruped Locomotion

Velocity tracking is a fundamental task in reinforcement learning for quadruped locomotion. Conventional velocity-tracking rewards are typically formulated in terms of the squared error between commanded and actual velocities. In this study, we propose a velocity-tracking reward from the controllability Gramian of cent...

Dynamic Monitoring and Emergency Collaborative Decision-Making for Logistics Supply Chain Risks Integrating GD-AHP and MADRL

Since there are various uncertain influences that affect the operation of the logistics supply chain, the existing supervision method cannot offer timely and efficient historical data, while the supply-demand relationship still needs enhancement. Therefore, this study puts forward a logistics supply chain risk dynamic...

On "Goal-Driven Life Forms": The Nature, Boundaries, and Safety Implications of AI

Abstract Contemporary large language models (LLMs) and reinforcement learning (RL) agents are inherently goal-seeking organisms: they optimize reward signals relentlessly, with no inherent ethical or epistemic constraints. While this intrinsic drive enables superhuman exploratory capability, it inevitably gives rise to...

THE EL-RAKHAWI FRAMEWORK FOR COGNITIVE TOPOLOGICAL METAMATERIALS: AI-DRIVEN PROGRAMMABLE MATTER FOR ZERO-LOSS ENERGY AUTONOMY AND EXTREME ENVIRONMENT ADAPTATION

The El-Rakhawi Framework for Cognitive Topological Metamaterials (CTM) introduces a revolutionary paradigm transforming passive matter into intelligent, adaptive systems. Addressing the fundamental limitation of static materials in extreme environments, CTM integrates higher-order topological insulators (HOTIs) with ph...

Autonomous Healthcare Control System (AHCS)

Today's healthcare systems represent complex adaptive systems that involve stochastic demand, tight coupling between subsystems, and nonlinearity of resource constraints. This paper proposes the Autonomous Healthcare Control System (AHCS), which brings together such approaches as stochastic state space models, Markov D...

Deep reinforcement learning-based Dynamic routing for ultra-reliable low-latency communication in 6G networks

Ultra-reliable low-latency communication (URLLC) is one of the hardest service classes that 6G has to carry, because a packet that misses its deadline is useless even if it eventually arrives. Routing based on fixed shortest paths ignores queue build-up, and link-state protocols that rebuild their tables every few mill...

Adaptive Self–Mutual Error Correction in World Models: A Unified Framework for State-Dependent Switching Between Internal Repair and Cross-Model Correction

World models provide internal representations of environmental dynamics and support prediction, planning, simulation, and decision-making. However, errors in world models may persist when a system relies exclusively on its own internal representations, while external correction can introduce communication costs, confli...