Chi-Chang Lee

Since July 2023, I have also been collaborating with the Improbable AI Lab, led by Prof. Pulkit Agrawal at MIT CSAIL. I earned both my Bachelor’s and Master’s degrees in Engineering Science and Computer Science from National Taiwan University, where I worked closely with Prof. Yu Tsao and Prof. Hsin-Min Wang.

Research: Before 2023, my research primarily focused on task-aware front-end processing for downstream audio applications, with a particular emphasis on developing robust speech-to-text systems (References: ICLR'23 and ASRU'23). Consequently, in 2023, I started my research focus to sensorimotor learning, especially Reinforcement Learning (RL) (References: ICRA'24 and NeurIPS'24). My current works now center on Optimal-invariant policy learning under local approximation (finite sample) constraints and Building connection in perceivability and decision-making. I am actively seeking a Ph.D. position starting in Fall 2025! Please feel free to reach out!

Experience

	Research Collaborator Jul. 2023 – present Improbable AI Lab at Massachusetts Institute of Technology Advisor: Pulkit Agrawal
	Research Assistant Mar. 2019 – Mar. 2024 Bio-ASP Lab at Academia Sinica CITI, Taiwan Advisor: Yu Tsao
	Visiting Researcher Nov. 2022 – Feb. 2023 Yamagishi Laboratory at National Institute of Informatics, Japan Advisor: Prof. Junichi Yamagishi

Selected Publications in Sensorimotor Learning
(∗ indicates equal contribution)

Optimal-invariant policy learning under local approximation (finite sample) constraints - In practice, deep RL training frequently relies on tedious heuristic reward designs to promote survival and exploration in complex environments, mainly due to the limitations of finite sample sizes in each training batch. My recent works tackle the exploitation-exploration dilemma under finite sample constraints, facilitating more practical approximations of optimal policies while enhancing the robustness of heuristic selections, all with significantly reduced design effort.

Going Beyond Heuristics by Imposing Policy Improvement as a Constraint
Chi-Chang Lee*, Zhang-Wei Hong*, Pulkit Agrawal
Conference on Neural Information Processing Systems (NeurIPS) 2024
Paper | OpenReview | Video | Code

Maximizing Velocity by Minimizing Energy
Srinath Mahankali*, Chi-Chang Lee*, Gabriel B. Margolis, Zhang-Wei Hong, Pulkit Agrawal
International Conference on Robotics and Automation (ICRA), 2024
Paper | Website | Code

Building connection in perceivability and decision-making - An agent cannot effectively make decisions without perceiving the necessary patterns. While trajectories for policy training are collected by experts using privileged information, but only sensor features (partial observations) are available during deployment, reproducing the same decision-making outcomes becomes challenging due to insufficient patterns in these partial observations. My earlier work focused on optimizing perception model training to ensure it aligns with both perception and downstream objectives.

D4AM: A General Denoising Framework for Downstream Acoustic Models
Chi-Chang Lee, Yu Tsao, Hsin-Min Wang, Chu-Song Chen
International Conference on Learning Representations (ICLR), 2023
Paper | Code

Publications in Audio Applications

LC4SV: A Denoising Framework Learning to Compensate for Unseen Speaker Verification Models
Chi-Chang Lee, Hong-Wei Chen, Chu-Song Chen, Hsin-Min Wang, Tsung-Te Liu, Yu Tsao
IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), 2023
Paper | Code

NASTAR: Noise Adaptive Speech Enhancement with Target-Conditional Resampling
Chi-Chang Lee, Cheng-Hung Hu, Yu-Chen Lin, Chu-Song Chen, Hsin-Min Wang, Yu Tsao
INTERSPEECH, 2022
Paper | Website | Code

SERIL: Noise adaptive speech enhancement using regularization-based incremental learning
Chi-Chang Lee, Yu-Chen Lin, Hsuan-Tien Lin, Hsin-Min Wang, Yu Tsao
INTERSPEECH, 2020
Paper | Code

Teaching

Teaching Assistant, Machine Learning, National Taiwan University, Taiwan 2021 Fall

Teaching Assistant, Time Frequency Analysis and Wavelet Transforms, National Taiwan University, Taiwan 2018 Fall

Honor

Second Place, IC/CAD Contest 2019

template from jonbarron