Scholar
Akifumi Wachi
Google Scholar ID: iC2b9GUAAAAJ
Senior Chief Research Scientist, LY Corporation
Citations & Impact
Citations
818
H-index
11
i10-index
12
Publications
20
Co-authors
31
list available
Publications
Resume (English only)
Academic Achievements
- - Publications:
- - Stepwise Alignment for Constrained Language Model Policy Optimizations, NeurIPS, 2024
- - Safe Exploration in Reinforcement Learning: A Generalized Formulation and Algorithms, NeurIPS, 2023
- - Safe Policy Optimization with Local Generalized Linear Function Approximations, NeurIPS, 2021
- - Safe Reinforcement Learning in Constrained Markov Decision Processes, ICML, 2020
- - Failure-Scenario Maker for Rule-Based Agent using Multi-agent Adversarial Reinforcement Learning and its Application to Autonomous Driving, IJCAI, 2019
- - Safe Exploration and Optimization of Constrained MDPs using Gaussian Processes, AAAI, 2018
Research Experience
- - May 2025 - present: Senior Chief Research Scientist, LY Corporation
- - Oct 2023 - Apr 2025: Chief Research Scientist, LY Corporation
- - Sep 2022 - Sep 2022: Senior Research Scientist, LINE Corporation
- - Apr 2018 - Aug 2022: Research Scientist, IBM Research AI
Education
- - Ph.D. in Computer Science, University of Tsukuba, 2021
- - M.S. in Aeronautics and Astronautics, University of Tokyo, 2018
- - B.S. in Aeronautics and Astronautics, University of Tokyo, 2016
Background
- - Research Interests: reinforcement learning (RL), reliable sequential decision making, AI Safety
- - Professional Fields: safe RL, RL × NLP, adversarial testing via RL
- - Brief Introduction: From Japan, currently a Senior Chief Research Scientist at LY Corporation, focusing on how policies should and can be trained and deployed in safety-critical problems.
Miscellany
- Personal interests and other information not provided
