GAIPS Lab
Research Group on AI for People and Society, INESC-ID
Email
Email
Facebook
Facebook
Instagram
Instagram
LinkedIn
LinkedIn
X
X
YouTube
YouTube
✨ Trimonthly Digest ✨ Oct, Nov & Dec 2025
✨ Trimonthly Digest ✨ Oct, Nov & Dec 2025
Check out this Document (A4) designed by Ana Carrasco.
✨ Our Website
✨ Our Website
Talks @ GAIPS: Astrid Rosenthal-von der Pütten
Talks @ GAIPS: Astrid Rosenthal-von der Pütten
Zoom is the leader in modern enterprise cloud communications.
📖 LisHAB - A Housing Market Agent-Based Simulation Model | EPIA'2025
📖 LisHAB - A Housing Market Agent-Based Simulation Model | EPIA'2025
With the recent price increases, housing affordability has become a critical issue in Portugal. This situation requires the right policies to be taken. To support informed policymaking, this work presents an agent-based model of the Lisbon Metropolitan Area housing...
📖 Optimizing 2D Packing Strategies for Autoclave Loading Using Deep Reinforcement Learning | EPIA'2025
📖 Optimizing 2D Packing Strategies for Autoclave Loading Using Deep Reinforcement Learning | EPIA'2025
📖 Centralized training with hybrid execution in multi-agent reinforcement learning via predictive observation imputation | ScienceDirect
📖 Centralized training with hybrid execution in multi-agent reinforcement learning via predictive observation imputation | ScienceDirect
We study hybrid execution in multi-agent reinforcement learning (MARL), a paradigm where agents aim to complete cooperative tasks with arbitrary commu…
📖 Robots with an Agenda foster Uncooperative Behaviors | ROMAN'2025
📖 Robots with an Agenda foster Uncooperative Behaviors | ROMAN'2025
📖 Regularization and Two Time Scales for Convergence of Reinforcement Learning | Applied Mathematics & Optimization in Springer Nature
📖 Regularization and Two Time Scales for Convergence of Reinforcement Learning | Applied Mathematics & Optimization in Springer Nature
Reinforcement learning algorithms aim at solving discrete time stochastic control problems with unknown underlying dynamical systems by an iterative process of interaction. The process is formalized as a Markov decision process, where at each time step, a control action is given, the system provides a reward, and the state changes stochastically. The objective of the controller is the expected sum of rewards obtained throughout the interaction. When the set of states and or actions is large, it is necessary to use some form of function approximation. But even if the function approximation set is simply a linear span of fixed features, the reinforcement learning algorithms may diverge. In this work, we propose and analyze regularized two-time-scale variations of the algorithms, and prove that they are guaranteed to converge almost-surely to a unique solution to the reinforcement learning problem.
📖 The Alien Bar Game: Puzzling Social Interaction Between Baristas
📖 The Alien Bar Game: Puzzling Social Interaction Between Baristas
📖 Exploring Compatible Interaction Preferences with a PuzzleVideo Game | DiGRA Digital Library
📖 Exploring Compatible Interaction Preferences with a PuzzleVideo Game | DiGRA Digital Library
View on mobile
