Hello World!

Hi, I’m Duo Zhou. I’m looking for a Ph.D. in CS/ECE/Information Science.

My research interests lie broadly within AI Safety, including Theory & Application of:

1. Trustworthy machine learning, especially Robustness (learning with non-i.i.d. data [1] or distribution shift [2]) and Adversarial Robustness (neural network verification [3] and adversarial training).

2. Sequential decision making under uncertainty (robust and distributionally robust reinforcement learning, online learning, and optimal control [2]).

3. Other areas of trustworthy machine learning, such as robust graph learning, data privacy, and fairness.

Projects

Publications

[3] Duo Zhou, Christopher Brix, Grani A. Hanasusanto, Huan Zhang. "Scalable Neural Network Verification with Branch-and-bound Inferred Cutting Planes." In The 38th Annual Conference on Neural Information Processing Systems (NeurIPS 2024).

[2] Hyuk Park, Duo Zhou, Grani A. Hanasusanto, and Takashi Tanaka. "Distributionally Robust Path Integral Control." In 2024 American Control Conference (ACC), pp. 1164-1171. IEEE, 2024.

[1] Ruibing Jin*, Duo Zhou*, Min Wu, Xiaoli Li, and Zhenghua Chen. "An Adaptive and Dynamical Neural Network for Machine Remaining Useful Life Prediction." IEEE Transactions on Industrial Informatics 20, no. 2 (2023): 1093-1102.

(*Indicates these authors contributed equally to this work.)

Talks

Distributionally Robust Path Integral Control on The 25th International Symposium on Mathematical Programming (ISMP 2024), Montréal, Canada. July 23th. 2024.

Research Projects

(Current) Neural Multi-stage Optimization Algorithm.

(Current) Adversarial Attack based on Searching Tree.

(Current) Robust Reinforcement Learning.

(NeurIPS 2024) Scalable Cutting Plane Algorithm for Neural Network Verification.

(ACC 2024) Robust Model Predictive Path Integral (MPPI) Control

(TII 2023) Deformable NN for Out-of-distribution Data Prediction.

(Undergrad Project) Enhancing Data Quality Through Personalized Video Content. Advised by Prof. Roman Kuc. 2021.

Project Overview: This project focuses on integrating YOLO and CLIP to generate custom datasets from personal video content for retraining, research, and model enhancement. The goal is to improve the quality of data by leveraging elements within the videos to create a tailored dataset, thereby optimizing performance in specific applications.

Repo: https://github.com/Lemutisme/YOLO-CLIP-zero-shot-test

(Undergrad Thesis) Future Prices Properties and Predictabilities. Advised by Prof. Alexei Chekhlov. 2020-2021.

Project Overview: This project explores the performance of eight futures markets from China and the United States using a Trend-Following strategy. The study incorporates the Kelly formula to calculate real-time profit-loss ratios and winning rates for dynamic position adjustments. A traversing method with two variables was implemented in MATLAB to determine optimal numerical values for the process. The results include comparisons between in-sample and out-of-sample performance across markets.

Repo: https://github.com/Lemutisme/Prediction-of-Future-Market

(ICM 2020) EDP Prediction Based on GA-BP NN and Cultural Preservation Model Based on SEIR.

Repo: https://github.com/Lemutisme/ICM2020-F-review

(CUMCM 2019) The Optimization of Taxi Dispatching at Airports.

Repo: https://github.com/Lemutisme/National-MCM-2019-B-review

Recognition

The Winner for both Regular & Extended Track in VNN-COMP24. 2024.

Team member of alpha-beta-CROWN.

First-Class Scholarship for the Graduating Class, Jinan University. 2021.

Meritorious Winner of Mathematical Contest in Modeling and Interdisciplinary Contes in Modeling. 2020.

Leader, with team members Jing Song and Yimei Gu.

Provincial First Prize, Contemporary Undergraduate Mathematical Contest in Modeling. 2019.

Leader, with team members Maolin Dong and Weijian Zhang.

Experience

Agency for Science, Technology and Research (A*STAR), Singapore
Research Intern Supervised by:
Dr. Zhenghua Chen, Dr. Ruibing Jin
Jan. 2022 - July 2022

I worked as a research intern in the Machine Intellection department, Institute for Infocomm Research (I2R), where I developed methods like Dynamic CNN and Knowledge Distillation on the task of Machine Remaining Useful Life (RUL) prediction, to enhance the trustworthiness. Besides of that, my research focused on Transfer Learning & Domain Adaptation.

Chinese University of Hong Kong, (CUHK) Shenzhen
Intern Supervised by:
Dr. Haifeng Wu
May 2021 - July 2021

I worked as an intern in the Shenzhen Finance Institute (SFI), where I served in Data Analysis and Visualization roles for the City ESG Ratings Research project, responsible for Data Cleaning, Feature Engineering, and Visualization, and developing Analysis Workflows, utilized Python, MySQL, Tableau, and Power BI.

ESG Ratings for China’s Urban Development has been launched!

Education

2022-2025

University of Illinois, Urbana-Champaign

The Grainger College of Engineering

M.S. in Industrial Engineering

Advisor: Prof. Grani A. Hanasusanto

2021-2022

Nanyang Technological University, Singapore

School of Physical and Mathematical Science

M.S. in Analytics

Supervisor: Prof. Xiaoli Li

2017-2021

Jinan University

School of Electrical and Information Engineering

B.Eng. in Packaging Engineering

Miscellaneous

My favorite authors are Jean-Paul Sartre, Garcia Marquez, and Xin Qiji.
I have a cute cat named Bailu (白露).