reinforcement learning

Prequential posteriors

Abstract Data assimilation is a fundamental task in updating forecasting models upon observing new data, with applications ranging from weather prediction to online reinforcement learning. Deep generative forecasting models (DGFMs) have shown excellent performance in these areas, but assimilating data into such models...

Verified Deep Learning with Lean 4

A formal verification of the math of modern deep learning, written as a Lean 4 blueprint project and shipped as the code it proves. Mathlib's fderiv is the foundation; every layer's backward is derived as a vector-Jacobian product rather than asserted -- dense, ReLU, softmax cross-entropy, 2D convolution, max-pool, Bat...

A Learnable Sparse Attention Graph Architecture for Heterogeneous Multi-UAV Air-to-Ground Mission Planning

This study provides an innovative and effective method for mission planning of heterogeneous multi-UAV systems in partially observable adversarial environments and outperforms the best-performing baseline algorithm by more than 20% across all operational performance metrics.

Reproduction code and screened corpus for a survey of evaluation design in reinforcement learning-based route guidance

Reproduction material for the survey "Study-Area Boundaries in Reinforcement Learning-Based Route Guidance: A Survey of Evaluation Design and ReportingRequirements" by Donghoun Lee. The archive contains the grid, benchmark and SUMO environments, the planner, the training code, the screened corpus withthe recorded field...

Reinforcement Learning for Game Playing

Progress in educational theory and translational implementation of simulation-based education for healthcare professionals: an integrative analysis from theory to effectiveness

This review offers a practical selection matrix and implementation governance model that assists educators in designing scalable and equitable SBE, supports administrators in resource planning and quality assurance, and guides accreditors in aligning evidence with entrustment and standards.

Supplemental Materials for "Uncertainty-aware Construction Task Scheduling for Human-robot Collaboration via Fuzzy-time Model and Deep Reinforcement Learning"

This record contains supplemental materials associated with the manuscript entitled “Uncertainty-aware Construction Task Scheduling for Human-robot Collaboration via Fuzzy-time Model and Deep Reinforcement Learning.” The supplemental materials provide extended figures and experimental results supporting the analyses pr...

OPTICAT: A Deep Reinforcement Learning Framework for Instance-Specific Algorithm Configuration

Optimization algorithms contain parameters that greatly influence their behavior. Finding the right settings for parameters through automated algorithm configuration has become a critical component of designing competitive algorithms. While traditional offline configurators tackle this problem by finding one configurat...

THE ROLE OF ARTIFICIAL INTELLIGENCE IN DIABETES CARE AND MANAGEMENT: A COMPREHENSIVE REVIEW

Diabetes mellitus represents a global metabolic crisis characterized by complex self-management requirements, marked physiological variability, and significant risks of microvascular and macrovascular complications. The integration of Artificial Intelligence (AI)—including Machine Learning (ML), Deep Learning (DL), Nat...

Adaptive bus bridging strategies for disrupted light rail transit networks via a two-stage reinforcement learning driven approach

Verified Deep Learning with Lean 4

A formal verification of the math of modern deep learning, written as a Lean 4 blueprint project and shipped as the code it proves. Mathlib's fderiv is the foundation; every layer's backward is derived as a vector-Jacobian product rather than asserted -- dense, ReLU, softmax cross-entropy, 2D convolution, max-pool, Bat...

Reproducibility package for "Formulation-holdout validation for reliable machine learning in replicated materials experiments: an AA6061 composite case study"

This record contains the complete reproducibility package supporting the manuscript "Formulation-holdout validation for reliable machine learning in replicated materials experiments: an AA6061 composite case study." The study evaluates 12 independently prepared AA6061 hybrid-composite specimens nested within four reinf...