L'Analyse Longue — Vie artificielle & calcul évolutionnaire

Helena et Clara discutent des actualités du jour.

Sources :

• arXiv cs.NE: The Causally Emergent Alignment Hypothesis: Causal Emergence Aligns with and Predicts Final Reward in Reinforcement Learning Agents
• arXiv cs.NE: How Information Evolves: Stability-Driven Assembly and the Emergence of a Natural Genetic Algorithm
• arXiv cs.NE: Adversarial Coevolutionary Illumination with Generational Adversarial MAP-Elites
• arXiv cs.NE: Quantization robustness from dense representations of sparse functions in high-capacity kernel associative memory