my research

On a mission to make AI more accessible and streamline AI development, I have been researching and developing MLOps infra. I am also interested in using AI to solve real-world problems, such as ads understanding, video analysis, and job-resume matching.

• I have published 20 papers in top-tier conferences, journals and workshop, including NeurIPS, ACMMM, ICDM, CIKM, ICML workshop, etc.
• I have also served as a reviewer for top-tier conferences and journals

LLM Agent for MLOps

I am building a large language model (LLM) agent to make MLEs' life easier.

• MLE Agent is a copilot for MLEs, which can help MLEs to write, reproduce and optimize ML code.
• AI Compliance Agent is a copilot for MLEs and Compliance officers. It can help them detect and fix compliance issues in their ML projects.

MLOps Infra

I have been researching and developing MLOps infra to streamline AI development.

• MLModelCI: An Automatic Cloud Platform for Efficient MLaaSHuaizheng Zhang, Yuanming Li, Yizheng Huang, Yonggang Wen, Jianxiong Yin, Kyle Guan, ACMMM 2020 (OSC)
• ModelCI-e: Enabling Continual Learning in Deep Learning Serving Systems Yizheng Huang, Huaizheng Zhang, Yonggang Wen, Peng Sun, Nguyen Binh Duong Ta, Tech Report
• InferBench: Understanding Deep Learning Inference Serving with an Automatic Benchmarking System Huaizheng Zhang, Yizheng Huang, Yonggang Wen, Jianxiong Yin, Kyle Guan, Tech Report
• Active-Learning-as-a-Service: An Automatic and Efficient MLOps System for Data-Centric AI Yizheng Huang, Huaizheng Zhang, Yuanming Li, Chiew Tong Lau, Yang You, NeurIPS 2022 Workshop on Human in the Loop Learning

ML-based Video Analysis

I used AI to build video analysis systems to optimize video quality of experience (QoE) and viewer interest inference. I also build serving systems to support more efficient video analytics.

• DeepQoE: A multimodal learning framework for video quality of experience (QoE) prediction Huaizheng Zhang, Linsen Dong, Guanyu Gao, Han Hu, Yonggang Wen, Kyle Guan, TMM
• A Serverless Cloud-Fog Platform for DNN-Based Video Analytics with Incremental Learning Huaizheng Zhang, Meng Shen, Yizheng Huang, Yonggang Wen, Yong Luo, Guanyu Gao, Kyle Guan, Tech Report
• Hysia: Serving DNN-Based Video-to-Retail Applications in Cloud Huaizheng Zhang, Yuanming Li, Qiming Ai, Yong Luo, Yonggang Wen, Yichao Jin, Nguyen Binh Duong Ta, ACMMM 2020 (OSC)
• Optimizing quality of experience for adaptive bitrate streaming via viewer interest inference Guanyu Gao, Huaizheng Zhang, Han Hu, Yonggang Wen, Jianfei Cai, Chong Luo, Wenjun Zeng, TMM
• Spatial-Temporal Federated Learning for Lifelong Person Re-identification on Distributed Edges Lei Zhang, Guanyu Gao, Huaizheng Zhang, TCSVT
• Towards Data-Efficient Continuous Learning for Edge Video Analytics via Smart Caching Lei Zhang, Guanyu Gao, Huaizheng Zhang, ACM SenSys 2022

AI Applications (Ads, Resume)

I have been researching and developing MLOps infra to streamline AI development.

• Look, Read and Feel: Benchmarking Ads Understanding with Multimodal Multitask Learning Huaizheng Zhang, Yong Luo, Qiming Ai, Yonggang Wen, ACMMM 2020
• ResumeGAN: An Optimized Deep Representation Learning Framework for Talent-Job Fit via Adversarial Learning Yong Luo*, Huaizheng Zhang*, Yonggang Wen, Xinwen Zhang, (*equal contributions), CIKM2019
• ResumeNet: A Learning-based Framework for Automatic Resume Quality Assessment Yong Luo*, Huaizheng Zhang*, Yongjie Wang, Yonggang Wen, Xinwen Zhang, (*equal contributions) ICDM 2018