Max Kleiman-Weiner @maxkw.bsky.social · 09/08/2024Through simulations, we demonstrate that an ISR model prevents agents from unlearning socialized behaviors and enables them to generalize to new tasks. The ISR model continues to guide the agent even when the caregiver is no longer present. 100
Max Kleiman-Weiner @maxkw.bsky.social · 09/08/2024Finally, we explore the effects of incomplete internalization and show "reward hacking," where the agent's behavior diverges from the intended goals. 100
Max Kleiman-Weiner @maxkw.bsky.social · 09/08/2024Check out the full manuscript here: arxiv.org/abs/2407.14681 and code: github.com/friedeggs/social-playarxiv.orgValue Internalization: Learning and Generalizing from Soc...Social rewards shape human behavior. During development, ... 000
