reinforcement learning

A risk-aware multimodal temporal reinforcement learning approach for collision avoidance of unmanned surface vehicles in multi-dynamic-obstacle scenarios

This study addresses autonomous collision avoidance for unmanned surface vehicles in multi-dynamic-obstacle maritime scenarios, where the number of obstacles varies, risk interactions are complex, and dynamic environmental evolution is difficult to exploit effectively. A Risk-Aware Multimodal Temporal Proximal Policy O...

Fábrica de Modelos — herramienta IA para la academia (UMH)

Recorrido interactivo por las siete etapas de construcción de un modelo de lenguaje —tokenización BPE, atención, preentrenamiento, ajuste a instrucciones, refuerzo e inferencia— con un Transformer que se entrena de verdad en JavaScript dentro de la propia página, sin instalar nada. Material didáctico construido sobre e...

AGENTIC SOCIO-TECHNICAL SYSTEMS FOR AGING DEMOGRAPHICS: INTEGRATING GRAPH NEURAL NETWORKS AND DEEP REINFORCEMENT LEARNING FOR RESILIENT HEALTHCARE LOGISTICS

The fast-growing aging demographics in both developed and developing countries are exerting increasing stress on healthcare systems, especially those that experience high old-age dependency rates.Healthcare logistics models based on traditional methods tend to employ static and reactive decision-making strategies, whic...

Structured Episodic Memory (SEM)

Structured Episodic Memory (SEM) is an experimental non-neural agent architecture that combines overlapping state abstractions, associative value memories, eligibility-based temporal credit assignment, goal-conditioned contact control, and context-dependent strategy memory. We evaluate SEM in Pong using structured stat...

Integrating Formula 3B (Learn, Practice, Compete) with the Application GO Expert as a Blended Learning Model: A Conceptualization in the Tutoring Context

The digital transformation of education has driven tutoring institutions to shift from conventional face-to-face instruction toward an integrated learning approach leveraging technology. One such approach is Formula 3B, a learning model developed by the author and implemented at Ganesha Operation (GO), the leading and...

The Combination of RAG and RL: A Technological Breakthrough and Future Directions in Mathematical Reasoning with LLMs

Large language models (LLMs) face challenges in mathematical reasoning tasks, including logical disjunction, process hallucination, and a loss of creativity, which severely restricts the reliability and application of LLMs for this purpose. To tackle these issues, this paper applies secondary research, systematically e...

BeamNG Research SDK (BRSDK)

BRSDK is a deterministic, high-frequency telemetry framework and Python SDK designed to bridge the Simulation-to-Real (Sim2Real) transfer gap in autonomous vehicle engineering, reinforcement learning, and vehicle dynamics within BeamNG.drive and BeamNG.tech.

Risk-aware eVTOL path planning for urban fixed-altitude cruise: A multimodal fusion reinforcement learning approach

The complexity and uncertainty of urban environments create an urgent need for safety-enhanced path planning in the autonomous navigation of electric vertical take-off and landing (eVTOL) aircraft. However, existing methods often focus solely on binary collision outcomes or suffer from local optima entrapment, failing...

Bilevel Traffic Flow Control Model for Highway Network Based on Deep Reinforcement Learning Under Emergency Events

A Review of Robot Adaptive Control Driven by Deep Reinforcement Learning

This review aims to clarify the development path of DRL-driven robot adaptive control, offering reference for theory, algorithm design, and engineering practice and it is hoped that this survey will help new researchers quickly grasp the landscape of the field.

Interpretable and Robust Car-Following Planning via Causality-Driven Deep Reinforcement Learning

CoT-RL-HAR: Chain-of-Thought Reinforcement Learning for Few-Shot Time-Series Human Activity Recognition