Reinforcement Learning and Bandits Lab
 |
PhD Students
047 陳明宏 (MS: 09/2023–08/2025, PhD: 09/2025–present)
048 王廣達 (MS: 09/2022–08/2024, PhD: 09/2024–present; Co-advised by Prof. Wen-Chih Peng)
083 郭建廷 (MS: 09/2024–08/2026, PhD: 09/2026–present; Co-advised by Prof. I-Chen Wu and Prof. Ti-Rong Wu)
MS Students
066 陳奎元 (BS@ NYCU-CS)
Topic: Zero-Shot RL
067 張家祐 (BS@ NTHU-Science)
Topic: Policy Gradient
068 柯國勛 (BS@ NTHU-CS)
Topic: Test-Time Alignment
069 張立宜 (BS@ NCHU-EECS)
Topic: Model-Predictive Control
070 蔡皓宇 (BS@ NTHU-Math)
Topic: Zero-Shot RL
058 皮恩亞 (BS@ NCU-Math)
Topic: Cross-Domain RL
077 黃浚瑀 (BS@ NYCU-IEM)
Topic: RL for Diffusion Language Models
078 陳映弘 Adams Chen (BS@ University of British Columbia, Canada)
Topic: Stochastic Gradient Bandits
079 鄭芯薇 (BS@ NYCU-CS)
Topic: RL Pretraining for Restless Bandits
080 林之謙 (BS@ NSYSU-CS)
Topic: Zero-Shot RL
081 楊博聖 (BS@ NTHU-CS)
Topic: RL Pretraining for LLMs
082 陳清海 Nicholas Eugene (BS@ NYCU-CS)
Topic: Zero-Shot RL
Research Assistants
056 林楷傑 (BS@ NYCU-CS; Undergraduate Member: 08/2023–07/2025, RA: 08/2025–present)
075 馬楷翔 (BS@ NYCU-CS; RA: 08/2025–present)
076 吳泓諺 (BS@ NYCU-CS; Undergraduate Member: 08/2024–06/2026, RA: 07/2026–present)
090 吳建中 (BS@ FJU-CS; RA: 09/2026–present)
Undergraduate Students
Joining in 2020
Joining in 2021
Joining in 2022
043 楊竣傑
Joining in 2023
055 林佑家
056 林楷傑
Joining in 2024
059 徐嘉亨
060 黃佑得
Joining in 2025
065 廖漢軒
071 劉家琪
072 余逸翔
073 江品寬
074 戴翊宸
Joining in 2026
084 黃冠瑋
087 陳宥廷
088 游閔皓
089 蘇郁凱
PhD Alumni
001 洪偉 (now at MediaTek 聯發科技)
Dissertation: Action-Constrained Reinforcement Learning
002 洪鈺恆 (now Postdoctoral Researcher at Georgia Tech)
Dissertation: Learning to Optimize: Bandit Algorithms and Reinforcement Learning for Black-box Decision Problems
029 連云暄 (now Assistant Research Fellow at CITI, Acedemia Sinica 中研院資創中心助研究員; Co-advised by Prof. Yu-Shuen Wang)
Dissertation: Bridging the Gap in Reinforcement Learning: Robust Algorithms and Techniques for Practical Applications
030 何國豪 (now at MediaTek 聯發科技; Co-advised by Prof. I-Chen Wu)
Dissertation: A Study of Deep Reinforcement Learning for Human-Like Behavior Modeling and Scheduling Optimization
RA Alumni
063 許嘉喻 (now PhD student at University of Michigan; RA: 08/2024–06/2025)
064 秦紫頤 (now PhD student at Max Planck Institute; RA: 03/2025–07/2025)
MS Alumni
003 林峻立 (MS: 09/2019–07/2021, BS@NCU-CS, now in MediaTek)
Thesis: Frank-Wolfe Policy Optimization for Reinforcement Learning with Action Constraints
004 謝秉瑾 (MS: 09/2019–07/2021, BS@NCCU-CS, now in Inventec)
Thesis: Reinforced Few-Shot Acquisition Function Learning for Bayesian Optimization
005 李昕 (MS: 09/2019–03/2022, BS@NCKU-CS, now in PHISON)
Thesis: Accelerating Gaussian Process Regression via Meta-Learned Neural Processes: A Utility-Based MAML Approach
006 蘇信恩 (MS: 02/2021–09/2022, BS@NCTU-CS, BS+MS in 5 years, now in Google)
Thesis: Coordinate Ascent Policy Optimization
007 黃柏愷 (MS: 09/2020–09/2022, BS@NCKU-CS, now in Trend Micro Inc.)
Thesis: Q-Pensieve: Boosting Sample Efficiency of Multi-Objective RL Through Memory Sharing of Q-Snapshots
008 郭俊廷 (MS: 09/2020–10/2022, BS@NCTU-EE, now in MediaTek)
Thesis: Adaptive-UCB for Online Restless Bandits
009 黃睿宇 (MS: 09/2020–10/2022, BS@NCTU-IEM, now in MediaTek)
Thesis: Multi-Objective Time-Varying Bayesian Optimization
010 楊上萱 (MS: 09/2020–11/2022, BS@NCTU-CS, now in MediaTek)
Thesis: Variance-Reduced Frank-Wolfe Policy Optimization for Action-Constrained RL
011 歐陽良雋 (MS: 09/2020–12/2022, BS@ NCTU-EE, now in MediaTek)
Thesis: Robustifying Proximal Policy Optimization Against Noisy Critics
020 楊祐維 (MS: 09/2021–10/2023, BS@ NCTU-CS, now in MediaTek)
Thesis: Model Selection for Offline Model-Based RL via Bayesian Optimization
021 吳程畯 (MS: 09/2021–12/2023, BS@ NCHU-Applied Math)
Thesis: Learning From Expert Demonstrations With Incomplete Observations
022 朱文滔 (MS: 09/2021–03/2024, BS@ NCU-Math)
Thesis: Provably Convergent Mixture-of-Experts in Reinforcement Learning
031 陳彥儒 (MS: 09/2022–06/2024, BS@ NCTU-Applied Math, now in Synopsys)
Thesis: Accelerated Policy Gradient: On the Convergence Rates of Nesterov's Momentum for Reinforcement Learning
032 張千祐 (MS: 09/2022–06/2024, BS@ NCTU-CS)
Thesis: Learning From Demonstrative Experts of Diverse Preference
033 潘冠蓁 (MS: 09/2022–07/2024, BS@ NCTU-CS, now in TSMC)
Thesis: Survive on Planet Pandora: Robust Cross-Domain RL Under Distinct State-Action Representations
034 黃亭晅 (MS: 09/2022–07/2024, BS@ NCTU-IMF)
Thesis: Cross-Domain Knowledge Transfer via Preference Consistency
035 葉佳翰 (MS: 09/2022–01/2025, BS@ NCTU-CS, now in TronFuture)
Thesis: Action-Constrained Imitation Learning
036 詹昀銘 (MS: 09/2022–05/2025, BS@ NCTU-CS)
Thesis: Improving Anesthesia Simulators via Reinforcement Learning
044 陳盈圖 (MS: 09/2023–07/2025, BS@ NYCU-CS, now in Google)
Thesis: A Reward-Free Viewpoint on Multi-Objective Reinforcement Learning
045 陳騰睿 (MS: 09/2023–07/2025, BS@ NCCU-Applied Math)
Thesis: Plan2Align: Predictive Planning Based Test-Time Preference Alignment in Paragraph-Level Machine Translation
046 陳子安 (MS: 09/2023–10/2025, BS@ NYCU-CS)
Thesis: Plan2Cleanse: Test-Time Backdoor Detection and Mitigation via Monte-Carlo Planning in Deep Reinforcement Learning
051 朱立民 (MS: 09/2023–12/2025, BS@ NSYSU-Applied Math)
Thesis: Semi-Supervised Cross-Domain Imitation Learning
043 楊竣傑 (Undergraduate Member: 08/2023–08/2024, MS: 09/2024–06/2026, BS@ NYCU-CS, now in Realtek)
Thesis: From Reward-Free Representations to Preferences: Rethinking Offline Preference-Based Reinforcement Learning
050 林睿騰 (Undergraduate Member: 02/2023–08/2024, MS: 09/2024–08/2026, BS@ NYCU-CS)
Thesis: WhittleFormer: Scalable Neural Index Policies for Dynamic Counter-UAV Jamming
038 温柏萱 (Undergraduate Member: 09/2022–08/2024, MS: 09/2024–09/2026, BS@ NYCU-CS)
Thesis: Does Reasoning Really Make LLMs Safer? A Capability-Dependent Study under Adversarial Inference
057 林禹亨 (MS: 09/2024–10/2026, BS@ NYCU-CS)
Thesis: Choose Before You Decode: Latent Q-Guided Generation for Diffusion Language Models
049 吳秉澍 (Undergraduate Member: 02/2023–08/2024, MS: 09/2024–11/2026, BS@ NYCU-CS)
Thesis: Understanding Value-Based RL for Language Model Fine-Tuning: What Helps, What Hurts, and Why