معرفی
Serena Booth is an incoming Assistant Professor in Computer Science at Brown University. Previously, she served as an AAAS AI Policy Fellow in the U.S. Senate, advising the Senate Banking Committee on AI policy. She holds a PhD from MIT CSAIL (2023) and a BA from Harvard College (2016). Her research focuses on human-AI interaction, specification design for AI systems, and ethical AI practices. She also worked as an Associate Product Manager at Google, scaling ARCore to 100 million devices.
Her research explores how humans specify AI behaviors, assess system success, and mitigate misalignment risks. Key contributions include Bayes-TrEx (model transparency via Bayesian sampling) and RoCUS (robot controller understanding). Her work has been supported by NSF GRFP and MIT Presidential Fellowships. She advocates for science policy equity through MIT's Science Policy Initiative and co-founded initiatives to support women in computing (e.g., GW6 at MIT).
- Education: PhD MIT CSAIL (2023), BA Harvard College (2016)
- Awards: Rising Star in EECS, HRI Pioneer, NSF GRFP
- Key Areas: Reward design pitfalls, human-robot trust, ethical AI curriculum development
Her recent publications analyze reward function misdesign (AAAI 2023), human-AI teaching frameworks (HRI 2022), and feature attribution reliability (AAAI 2022). She currently seeks PhD students/postdocs focusing on human-AI alignment, reinforcement learning, and policy implications.
