4 papers
A Review on Single-Problem Multi-Attempt Heuristic Optimization
Judith Echevarrieta, Etor Arza, Aritz Pérez +1
In certain real-world optimization scenarios, practitioners are not interested in solving multiple problems but rather in finding the best solution to a single, specific problem. W…
Enabling Population-Based Architectures for Neural Combinatorial Optimization
Andoni Irazusta Garmendia, Josu Ceberio, Alexander Mendiburu
Neural Combinatorial Optimization (NCO) has mostly focused on learning policies, typically neural networks, that operate on a single candidate solution at a time, either by constru…
Self-Composing Policies for Scalable Continual Reinforcement Learning
Mikel Malagón, Josu Ceberio, Jose A. Lozano
This work introduces a growable and modular neural network architecture that naturally avoids catastrophic forgetting and interference in continual reinforcement learning. The stru…
Craftium: Bridging Flexibility and Efficiency for Rich 3D Single- and Multi-Agent Environments
Mikel Malagón, Josu Ceberio, Jose A. Lozano
Advances in large models, reinforcement learning, and open-endedness have accelerated progress toward autonomous agents that can learn and interact in the real world. To achieve th…