I’m a third year Artificial Intelligence PhD student at Berkeley working with Anca Dragan and Stuart Russell within BAIR and CHAI. I’m thankful to be supported by the NSF Fellowship.
I’m broadly interested in ensuring that AI systems interfacing with humans will lead to beneficial individual and societal outcomes.
I’m currently working on recommender systems, investigating how the choice of algorithm might affect us users. Another interest of mine is that of human-AI collaboration, with the goal of improving the quality and robustness of human models and of agents trained to collaborate with humans.
Before immigrating to the US, I grew up in the amazingly chaotic city of Livorno 🇮🇹 - visit if you get the chance!
- October 2022: "Uni[MASK]: Unified Inference in Sequential Decision Problems" was accepted as NeurIPS Oral.
- June 2022: I was accepted to be part of the first iteration of the AI Policy Hub.
- May-June 2022: I gave a talk at DeepMind, and at the Center for Human-Compatible AI Workshop: Coping with Recommender Mis-Alignment.
- April 2022: I was awarded the NSF Fellowship.
- March 2022: I gave a talk at MIT about our work Estimating and Penalizing Induced Preference Shifts in Recommender Systems.
Uni[MASK]: Unified Inference in Sequential Decision Problems
Micah Carroll, Orr Paradise, Jessy Lin, Raluca Georgescu, Mingfei Sun, David Bignell, Stephanie Milani, Katja Hofmann, Matthew Hausknecht, Anca Dragan, Sam Devlin. NeurIPS 2022 (Oral).
Estimating and Penalizing Induced Preference Shifts in Recommender Systems.
Micah Carroll, Dylan Hadfield-Menell, Stuart Russell, Anca Dragan. ICML 2022 (previous versions at Recsys 2021 LBR Track, and Recsys 2021 FAccTRec Workshop as long talk).
Optimal Behavior Prior: Improving Human-AI Collaboration Through Generalizable Human Models.
Mesut Yang, Micah Carroll, Anca Dragan. Human-in-the-loop Learning (HILL) Workshop, NeurIPS 2022.
Time-Efficient Reward Learning via Visually Assisted Cluster Ranking
David Zhang, Micah Carroll, Andreea Bobu, Anca Dragan. Human-in-the-loop Learning (HILL) Workshop, NeurIPS 2022.
Evaluating the Robustness of Collaborative Agents.
Paul Knott, Micah Carroll, Sam Devlin, Kamil Ciosek, Katja Hofmann, Anca Dragan, Rohin Shah. AAMAS 2021.
On the Utility of Learning about Humans for Human-AI Coordination.
Micah Carroll, Rohin Shah, Mark Ho, Tom Griffiths, Sanjit Seshia, Pieter Abbeel, Anca Dragan. NeurIPS 2019.
[video] [environment] [demo] [blogpost]