Assistant Professor City University of Hong Kong (CityU) luzhichaocn [at] gmail [dot] com Zhichao Lu received Ph.D degree in Electrical and Computer Engineering from Michigan State University under the supervision of Prof. Kalyanmoy Deb (FIEEE, FACM), where he studied bilevel optimization, evolutionary multi-objective optimization, and neural architecture search. In the broad context of AI, his current research focuses on the intersections of evolutionary computation, learning, and optimization, notably on developing efficient and automated ML/DL algorithms and systems, with the overarching goal of making AI accessible to everyone.
Prospective Students: I am always looking for self-motivated students with strong mathematical and programming background. Please refer to openings for potential opportunities.
- 06/2023: Tutorial on "Multi-Objective Optimization for Deep Learning" is accepted to CVPR 2023!
- 05/2023: One paper accepted to IEEE ICWS '23!
- 03/2023: Two papers accepted to CVPR '23! Congratulations to Shihua Huang, Chuntao Ding and all collaborators!
- 09/2021: Our HZL team won the Excellent Award (ranked 6/566) of Glaucoma grAding from Multi-Modality imAge (GAMMA) Challenge at the MICCAI OMIA workshop'21!
- 08/2021: Thanks National Natural Science Foundation of China to fund our surrogate-assisted evolutionary neural architecture search project (国家自然科学基金青年科学基金项目)!
- 07/2021: Two papers accepted to ICCV '21! Congratulations to Shihua Huang, Teng Wang and all collaborators!
- 06/2021: Our team is the Runner-Up Winner of the 2021 CVPR ActivityNet Event Dense-Captioning Challenge. Congratulations to Teng Wang and Zhu Liu for leading the project! Oral presentation at CVPR 21 Workshop.
- 06/2021: Thanks China Postdoctoral Science Foundation to fund our evolutionary multi-objective neural architecture search project (中国博士后科学基金面上资助项目)!
- 02/2021: Our paper titled "A New Many-Objective Evolutionary Algorithm Based on Generalized Pareto Dominance" is accepted to IEEE T-CYB! Congratulations to Shuwei Zhu for leading the paper!
- 01/2021: Our paper titled "Neural Architecture Transfer" is accepted to IEEE T-PAMI!
- 09/2020: Our paper titled "Multiobjective Evolutionary Design of Deep Convolutional Neural Networks for Image Classification" is accepted to IEEE T-EVC!
- 07/2020: NSGANetV2 is accepted as Oral (top 2%) at ECCV '20!
- 06/2020: NSGA-Net is invited to IJCAI '20 as an extended abstract.
- 04/2020: MUXConv is accepted to CVPR '20!
- 07/2019: NSGA-Net won the Best Paper Award at GECCO '19 (Evolutionary Machine Learning Track).
|
Zhichao Lu , Ran Cheng, Yaochu Jin, Kay Chen Tan, Kalyanmoy Deb IEEE TEVC, 2023 arXiv / slides / code Transforming Neural Architecture Search (NAS) into multiobjective optimization problems. A benchmark suite for testing evolutionary algorithms in deep learning. |
|
Shihua Huang, Zhichao Lu , Kalyanmoy Deb, Vishnu N. Boddeti CVPR, 2023 arXiv / poster / code A holistic study on the impact of architectural design on adversarial robustness, showing that architecture contributes as much as, if not more than, adversarial training to adversarial robustness. |
|
Zhichao Lu , Gautam Sreekumar, Erik Goodman, Wolfgang Banzhaf, Kalyanmoy Deb, Vishnu N. Boddeti IEEE TPAMI, 2021 arXiv / video / code Improve practical utilities of neural architecture search through many-objective optimization, iterative surrogate modeling, and transfer learning. |
|
Zhichao Lu , Kalyanmoy Deb, Erik Goodman, Wolfgang Banzhaf, Vishnu N. Boddeti ECCV, 2020 (Oral Presentation) arXiv / video / code An efficient multi-objective neural architecture search framework. It can be easily integrated with most of search spaces. |
|
Zhichao Lu , Ian Whalen, Vishnu N. Boddeti, Yashesh Dhebar, Kalyanmoy Deb, Erik Goodman, Wolfgang Banzhaf GECCO, 2019 (Oral Presentation, Best Paper Award) arXiv / video / code / extended abstract (invited by IJCAI '20) An evolutionary multi-objective search algorithm for evolving DNN architectures automatically. |