Hima Lakkaraju 2023 Early Career Fellow
Affiliation Assistant Professor, Harvard University Hard Problem Solved AI’s continually evolving safety and security, robustness, performance, output challenges and other shortcomings that may cause harm or erode public trust of AI systems, especially in safety-critical applications and uses where societal stakes and potential for societal harm are high.

Himabindu (Hima) Lakkaraju is an Assistant Professor at Harvard University and her research spans the broad area of trustworthy machine learning. She focuses on improving the interpretability, fairness, privacy, robustness, and reasoning capabilities of various machine learning models including large language models and text-to-image models. Hima has co-authored more than 60 publications at top-tier machine learning conferences (e.g., ICML, NeurIPS, ICLR, AISTATS) as well as interdisciplinary journals (e.g., Nature Machine Intelligence, Nature Scientific Data), and has given keynote talks at various prestigious venues including the Stanford HAI conference, EmTech MIT conference, CIKM, ICML, NeurIPS, ICLR, AAAI, and CVPR. She has been named one of the world’s top innovators under 35 by both MIT Tech Review and Vanity Fair. She also received several prestigious awards including the National Science Foundation (NSF) CAREER award and multiple best paper awards at top-tier ML conferences, and grants from NSF, Google, Amazon, JP Morgan, and Bayer. Her research has been showcased by popular media outlets including the New York Times, TIME magazine, and Forbes. More recently, Hima co-founded the Trustworthy ML Initiative to enable easy access to resources on trustworthy machine learning, and to build a community of researchers and practitioners working on the topic.

AI2050 Project

Machine learning models are increasingly being used in high-stakes decision making across domains such as healthcare and law. In response, several regulatory principles have been established to guarantee their responsible use. A notable principle is the “Right to Explanation” (RTE), which mandates that individuals adversely affected by algorithmic decisions be given clear and actionable explanations. While there is abundant literature on implementing RTE in predictive models, its application to large language models remains under explored. Hima Lakkaraju’s AI2050 project seeks to address this gap, aligning policy and research to ensure that RTE is effectively enforced within the realm of large language models.