Nika Haghtalab 2023 Early Career Fellow
Affiliation Assistant Professor, UC Berkeley 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.

Nika Haghtalab is an Assistant Professor in the Department of Electrical Engineering and Computer Sciences at UC Berkeley. She works broadly on the theoretical aspects of machine learning and algorithmic economics. Prof. Haghtalab’s work builds theoretical foundations for ensuring both the performance of learning algorithms in the presence of everyday economic forces and the integrity of social and economic forces that are born out of the use of machine learning systems. Among her honors are the CMU School of Computer Science Dissertation Award, SIGecom Dissertation Honorable Mention, NeurIPS and ICAPS best paper awards, and an EC exemplary paper in the AI track award. Prof. Haghtalab is a co-founder of Learning Theory Alliance (LeT-All), a first large-scale mentoring activity for the theory of machine learning community.

AI2050 Project

Machine learning systems have become integral to our society, interacting with different sectors in unique and challenging ways. Yet, methods for addressing social and strategic considerations in ML are still in their infancy and lack provable performance. Nika Haghtalab’s AI2050 project aims to provide a unified perspective on a range of social and strategic considerations in ML, including robustness and fairness of learning systems.

Project Artifacts

D. Halawi, A. Wei, E. Wallace, T.T. Wang, N. Haghtalab, J. Steinhardt. Covert Malicious Finetuning: Challenges in Safeguarding LLM Adaptation. arXiv. 2024.