|
|
hotmail.com 
I am a machine learning researcher and software engineer currently working at VMWare.
My general research interests span machine learning, natural language processing, computational advertising, AIOps and compute games. Current focus of my work includes:
- Analyzing incident root causes from virtual machine log messages in cloud system
- Applying advance machine learning solutions in intelligent event management for VMWare clouds
 |
2020 - present | Member of Technical Staff, Cloud Management Team, San Francisco Bay Area CA, U.S. | |
 |
2019 - 2020 | Machine Learning Engineer, eMobility Team, San Francisco Bay Area CA, U.S. | |
 |
2017 summer | Intern, Traffic Science Team, Bellevue WA, U.S. | |
|
|
2016 summer | Intern, Traffic Science Team, Bellevue WA, U.S. | |
 |
2013 - 2019 | Ph.D, Computer Sicence, Oregon State University, U.S. (Thesis, Presentation) | |
|
|
2011 - 2013 | M.S., Computer Sicence, Oregon State University, U.S. | |
 |
2007 - 2011 | B.S., Computer Sicence, Tianjin University, China. | |
 |
2004 - 2007 | Tianjin Nankai High School, China. |
|
▪ Search-based Approaches for Natural Language Understanding Within Document Entity Coreference Resolution. Joint Entity-linking, Coreference and NER for Deep Language Understanding. HC-Networks for Search-based Structured Prediction. ▪ AIOps for Clouds Log-Based Root Cause Analysis for VMs in the Cloud System. Fault Event Classification in Intelligent Event Management. ▪ Evaluation Learning in Computer Chess My Computer Chess Program: Cmcchess. Linear Chess Evaluation Function Learning with Regression Learning. |
 |
Log-Based Root Cause Analysis at Scale | |
| ° Chao Ma, Darren Brown, Xing Wang, Junyuan Lin, Andrii Malinovskyi, Alexey Zavitaev, Nick Stephen | ||
| ° Conference of VMWare annual Research And Development Innovation Offsite (RADIO) 2022 | ||
|
Randomized Greedy Search for Structured Prediction: Amortized Inference and Learning [Appendix] [Slides] [Poster] [Code] | |
| ° Chao Ma, F A Rezaur Rahman Chowdhury, Aryan Deshwal, Md Rakibul Islam, Janardhan Rao Doppa, and Dan Roth | ||
| ° Proceedings of International Joint Conference on Artificial Intelligence (IJCAI) 2019 | ||
|
Joint Neural Entity Disambiguation with Output Space Search | |
| ° Hamed Shahbazi, Xiaoli Fern, Reza Ghaeini, Chao Ma, Rasha Obeidat, Prasad Tadepalli | ||
| ° Proceedings of International Conference on Computational Linguistics (COLING) 2018 | ||
|
Multi-Task Structured Prediction for Entity Analysis: Search-based Learning Algorithms [Supp] [Slides] [Poster] | |
| ° Chao Ma, Janardhan Rao Doppa, Prasad Tadepalli, Hamed Shahbazh, and Xiaoli Fern | ||
| ° Journal of Machine Learning Research (JMLR), Proceedings Track, Vol 77, 16 pages, 2017 | ||
|
Select-and-Evaluate: A Learning Framework for Large-Scale Knowledge Graph Search | |
| ° F A Rezaur Rahman Chowdhury, Chao Ma, Md Rakibul Islam, Mohammad Hossein Namaki, Mohammad Omar Faruk, and Janardhan Rao Doppa | ||
| ° Journal of Machine Learning Research (JMLR), Proceedings Track, Vol 77, 16 pages, 2017 | ||
|
Learning Greedy Policies for the Easy-First Framework [Slides] | |
| ° Jun Xie, Chao Ma, Janardhan Rao Doppa, Prashanth Mannem, Xiaoli Fern, Tom Dietterich, and Prasad Tadepalli | ||
| ° Proceedings of AAAI Conference on Artificial Intelligence (AAAI) 2015 | ||
|
Prune-and-Score: Learning for Greedy Coreference Resolution [Poster] [Code] | |
| ° Chao Ma, Janardhan Rao Doppa, Xiaoli Fern, Tom Dietterich, and Prasad Tadepalli | ||
| ° Proceedings of International Conference on Empirical Methods in Natural Language Processing (EMNLP), 2014 | ||
|
HC-Search for Multi-Label Prediction: An Empirical Study | |
| ° Janardhan Rao Doppa, Jun Yu, Chao Ma, Alan Fern, and Prasad Tadepalli | ||
| ° Proceedings of AAAI Conference on Artificial Intelligence (AAAI) 2014 |
- Tutorial: Recent Advances in Structured Prediction at AAAI2018 (with Janardhan Rao Doppa and Liping Liu [Slides]
 |
OregonStateCoref The coreference resolution resolver developed by Oregon State University Deep Reading and Learning research group. |
|
 |
Cute-MPI-Interface A program that allows a cutechess task to be able to run on a computer cluster system with MPICH2. A cutechess task usually involves hundreds or thousands of games. This small tool simply splits the task uniformly, and distributes them on different cluster nodes. |
|
 |
Senpai2-Regression A linear regression learner to train the evaluation weight vector of Senpai 2 with the NNUE training data. The current best-performed learned weight has a comparable strength against the original weight of Senpai under the 40moves/5mins time control. |
|
 |
C-Gomoku-Cli c-gomoku-cli is a command line interface for gomoku/renju engines that support Gomocup protocol. It provides an open-source tool that can run large scale engine-vs-engine matches on multi-core workstations or cluster systems in parallel. |
|