Rachel Freedman
Papers
1
Total Citations
4
H-Index
1
About
Rachel Freedman is a leading researcher in the intersection of artificial intelligence, robotics, and value alignment, with a primary focus on ensuring that autonomous systems behave safely and ethically. Her most influential work addresses the critical challenge of reward inference, particularly the problem of choice set misspecification—where an incorrectly defined set of possible actions can lead an AI to infer flawed human preferences and pursue dangerous behaviors. This foundational paper, "Choice Set Misspecification in Reward Inference" (2021, with 4 citations), has already shaped how researchers think about the subtle pitfalls in learning reward functions from human feedback. Freedman’s contributions are vital for developing robots that can operate reliably in complex, real-world environments where explicit reward signals are absent. Her work bridges technical rigor with deep philosophical questions about human-AI interaction, making her a key voice in the safe AI community. Through her research, Freedman is helping to build the theoretical and practical foundations for trustworthy autonomous systems that truly understand and align with human values.
Research Focus
Key Achievements
Top Papers
- 1Choice Set Misspecification in Reward Inference4 citations · 2021