Program

Workshop opening

Matteo Cederle, Marco Fabris and Gian Antonio Susto: A Distributed Approach to Autonomous Intersection Management via Multi-Agent Reinforcement Learning

Mladen Miletić, Ivana Dusparic and Edouard Ivanjko: Growing Neural Gas in Multi-Agent Reinforcement Learning Adaptive Traffic Signal Control

Ana L. C. Bazzan and Henrique Gobbi: Learning From Your Virtual Neighbors: a Reinforcement Learning Approach to Traffic Signal Control

Coffee Break

Davide Tosi, Riccardo Cappi and Sebastiano Monti: Deep convolutional Q-learning for traffic lights optimization in Smart Cities

Anastasios Kontogiorgis and Melanie Bouroche: CAV Unsignalised Intersection Crossing with Unreliable Emergent Communication

Shuncheng Cai and Mélanie Bouroche: Multi-Agent Deep Reinforcement Learning for Longitudinal Control of CAVs in Mixed Traffic Environments with Unreliable Communication

Bharathkumar Hegde and Mélanie Bouroche: Safe CAV lane changes using MARL and control barrier functions

Lunch

Elif Arslan, Emir Demirović, Marco Rinaldi, Oded Cats and Serge Hoogendoorn: Explainable Demand Prediction for Ride-Hailing Services

Andre Artelt and Andreas Gregoriades: Application of Multi-Instance Counterfactual Explanation in Road Safety Analysis

Patrick O'Hara, Ramanujan M. Sridharan and Theo Damoulas: Routing on Sparse Graphs with Non-metric Costs for the Prize-collecting Travelling Salesperson Problem

Marvin Damschen, Rickard Häll, Anders Thorsén and Ashfaq Farooqui: Assessing a UAS for Maritime Firefighting and Rescue on Ro-Ro Ships

Coffee Break

Giuseppe Vizzari, Daniela Briola and Federico Pisapia: Curriculum-Based RL for Pedestrian Simulation: Sensitivity Analysis and Hyperparameter Exploration

Mathieu Bourgais, Arnaud Saval, Pierrick Tranouez, Olivier Gillet and Eric Daudé: Designing a Serious Game of crisis management on top of an Agent-Based Simulation of population evacuation

Alaa Daoud, Corentin Bunel and Maxime Guériau: Introducing Multiagent Systems for AV Visual Perception Sub-tasks: A proof-of-concept implementation for bounding-box improvement

Workshop closing