Scholar
Hyunin Lee
Google Scholar ID: kHTDu1YAAAAJ
University of California, Berkeley
Citations & Impact
Citations
86
H-index
4
i10-index
2
Publications
9
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- - Publications:
- * New paper on Multi-Modal Learning: Orthogonal Alignment, to be presented in ICML2025
- * Extension of Black Swan paper: Antifragile, to be presented in ICML2025
- * A Black Swan Hypothesis: The Role of Human Irrationality in AI Safety, oral presentation at ICLR2025 workshop (Financial AI)
- * Policy-based Primal-Dual Methods for Concave CMDP with Variance Reduction
- * Pausing Policy Learning in Non-stationary Reinforcement Learning (ICML 2024, Oral, top 1%)
- * Beyond Exact Gradients: Convergence of Stochastic Soft-Max Policy Gradient Methods with Entropy Regularization (IEEE TAC)
- * Tempo Adaption in Non-stationary Reinforcement Learning (NeurIPS 2023)
- * Initial State Interventions for Deconfounded Imitation Learning (CDC 2023)
- * Explainable Deep Learning Model for EMG-Based Finger Angle Estimation using Attention (IEEE TNSRE)
- - Awards:
- * Berkeley research fellowship, 2024
- * NeurIPS scholar award, 2023
- * Kwanjeong Education Foundation Scholarship, 2022
- - Projects:
- * Working with OUTTA to create intuitive AI educational materials
- - Reviewer Experience:
- * Reviewer for ICML, RLC, and Neurips 2025
- * Reviewer for ICLR, AISTATS 2025
- * Reviewer for ICML 2024
Research Experience
- - Research Scientist Intern (Ranking team) at Meta, Summer 2025
- - Research Associate at OpenAI, started in March 2025
- - Co-founded and leading the AI PLAYGROUND team, dedicated to creating intuitive AI educational materials for elementary education, launched in October 2023
Education
- - Ph.D.: University of California, Berkeley, 2022.08 - Present, Advisor: Somayeh Sojoudi
- - B.S.: Seoul National University, 2015.03 - 2022.02, Advisor: Yong-Lae Park (Served military service in Korean Combat Training Center, 2017.07 - 2019.03)
Background
- - Research Interests: Reinforcement Learning, Multi-Modal Learning
- - Professional Field: Developing reinforcement learning algorithms and exploring their key concepts
- - About: Fourth-year Ph.D. student at UC Berkeley, interested in AI education, co-founded and currently advising OUTTA
Miscellany
- - Interested in AI education, co-founded and currently advising OUTTA
