3 papers
cs.AI2025
HAVA: Hybrid Approach to Value-Alignment through Reward Weighing for Reinforcement Learning
Kryspin Varys, Federico Cerutti, Adam Sobey +1
Our society is governed by a set of norms which together bring about the values we cherish such as safety, fairness or trustworthiness. The goal of value-alignment is to create age…
cs.LG2024
CHIRPs: Change-Induced Regret Proxy metrics for Lifelong Reinforcement Learning
John Birkbeck, Adam Sobey, Federico Cerutti +2
Reinforcement learning (RL) agents are costly to train and fragile to environmental changes. They often perform poorly when there are many changing tasks, prohibiting their widespr…
cs.CL2024
Speaking Your Language: Spatial Relationships in Interpretable Emergent Communication
Olaf Lipinski, Adam J. Sobey, Federico Cerutti +1
Effective communication requires the ability to refer to specific parts of an observation in relation to others. While emergent communication literature shows success in developing…