reinforcement learning

Calibration-Aware Reinforcement Learning for Large Language Models: A Survey of Objectives, Optimization, and Decision-Making

Large language models increasingly produce confidence reports, predictive distributions, and structured decisions that determine whether a system answers, abstains, retrieves evidence, or spends additional computation. Reinforcement learning can improve these signals, but it can also change the answers being assessed,...

Navigating the Polycrisis

Developing economies are particularly prone to polycrisis risk given their inherent fragilities, volatile capital flows, and exposure to geopolitical events. The financial risk methodologies, such as value at risk modeling and linear econometric forecasting, cannot accommodate nonlinear risk interactions across differe...

Overcoming the Energy–Health Trade-Off in Smart Buildings: AI-Driven Dynamic Thresholding for Sustainable HVAC (Heating, Ventilation, and Air Conditioning) Actuation and Multi-Pollutant Management

An intelligent, energy-efficient, and robust multi-pollutant forecasting and control framework that integrates hybrid LSTM–GRU forecasting with Proximal Policy Optimization (PPO)-based reinforcement learning to maintain indoor air quality while minimizing unnecessary energy consumption and mechanical actuation is devel...

RegTrace v1 — FDA Regulatory Trajectory Dataset for Reinforcement Learning

This page on Zenodo hosts a free sample only (50 trajectories) — not the full dataset. The complete RegTrace v1 dataset — 29,395 trajectories, 457 SFT traces with full CRL letter text, and 795 forecasting records — is available at sentineldata.com.ua/dataset/regtrace-v1. This sample is a true, unmodified subset of the...

Hybrid AI Algorithm for Optimizing Radio Coverage in UAV-based Wireless Mesh Networks

This article introduces a hybrid GA–BO–DRL optimization framework designed to enhance ground node coverage in UAV enabled wireless mesh networks. The proposed methodology integrates a global search mechanism based on a genetic algorithm, a Bayesian optimization module for local refinement, and a deep reinforcement lear...

AI-Enabled Dynamic Pricing Models in E-Commerce Logistics

A deep reinforcement learning-based artificial intelligence framework of time-based e-commerce pricing is introduced, in which the dynamic pricing is modelled as a Markov Decision Process based on demand trends, inventory, user behaviour, and logistical restriction. The method presents a DRCR-based rewarding mechanism,...

Learning to Route On-Chip: A Survey of Machine Learning-Based Routing in Networks-on-Chip and Its Hardware Overheads

This survey reviews machine-learning-based NoC routing across a 48-study evidence base, and identifies three recurring gaps: limited scalability beyond small meshes, missing deadlock-freedom guarantees for learned policies, and over-reliance on synthetic traffic.

Supplementary Document for Origin-Destination-Aware ConstrainedGraph Reinforcement Learning for Dynamic Electric Vehicle Routingand Charging Coordination

A Hybrid Relay Admissibility and Security Weighting Architecture for Dynamic Adversarial Resilience in Wireless Sensor Networks

This repository contains the complete simulation engine, canonical results, statistical analysis pipeline, and verification/audit trail supporting the manuscript submitted to Scientific Reports. The archive includes: a custom discrete-event WSN simulator implementing a BIS-weighted relay admissibility filter combined w...

Graph-Enhanced Multimodal Knowledge Tracing with Reinforcement Learning for Explainable Personalized Music Learning Path Optimization

Personalized learning in AI-driven music education still faces challenges such as the difficulty of integrating multimodal learning evidence, the complexity of modeling music skill dependencies, and the lack of long-term optimization of learning pathways. To address these issues, this paper proposes a framework for opt...

Deep Reinforcement Learning for Tengiz-Sileno Supply Chain — Simulation Code and Results

This dataset contains the simulation code and results for the research paper "Deep Reinforcement Learning for Supply Chain Resilience: The Tengiz–Sileno Ethane Corridor in Kazakhstan." Contents: 1. Python source code for the physics-based simulator of the Tengiz → GSK → Sileno ethane supply chain (sileno_env.py, config...

Investigating Methods of Artificial Intelligence in Fiber Optic Network Performance Monitoring

Background: Elastic and cognitive optical networks vary modulation format, symbol rate, launch power and route while a service is carrying traffic, so the physical layer has to be measured continuously rather than engineered once at commissioning. Optical performance monitoring (OPM) and modulation format identificatio...