CMU FLAME Center Seminar - Hao Zhu

— 2:00pm

Location:
In Person and Virtual - ET - Tepper Building 1403

Speaker:
HAO ZHU , Postdoctoral Researcher, Computer Science Department, Stanford University
https://www.zhuhao.me/

Ushering AI Agents to an Open Social World

Unlike frontier AI models trained on static datasets, humans learn through dynamic interactions with other people and the world. This fundamental difference in learning methodology not only makes language agents less sample-efficient than humans but also introduces significant risks when these agents are deployed to interact with real humans in the real world. Building agents that can efficiently learn through interaction with other agents, humans and the world is a challenging problem. In this presentation, I will outline three foundational approaches we've developed to address this challenge:

  1. Learning through exploration on the internet (NNetNav-live) — We deploy an open-ended agent (without explicit task instructions) to explore the web, gather experience and retroactively label and train on the data.
  2. Learning from human normative decision-making (EgoNormia) — We explore methods for agents to observe and internalize social norms in physical interactions through crowd-sourced annotation with context perturbation.
  3. Learning to build metrics from human feedback (AutoLibra, in prep) — We present a framework for automatically building behavior evaluation metric systems that help both humans understand agent performance, and agents improve the policy based on human feedback.

These complementary approaches offer a path toward creating AI agents that can more effectively learn, adapt, and integrate into our open social world. 

— 

Hao Zhu is a postdoctoral researcher in the Computer Science Department at Stanford University. He finished his PhD last summer from CMU LTI. He is interested in AI agents, human-agent interaction, robotics and embodied AI, and what AI agents tell us about human social and embodied cognition. 

Pizza Lunch Provided

In Person and Zoom Participation.  See announcement.

Add event to Google
Add event to iCal