Zhiguang (Stephen) Wang

I am the Head of Engineering (US CTO) at Amber Group, a 3B Fintech unicorn. Before Amber, I am an engineering leader at Facebook Reality Labs (FRL). Before that, I worked as a researcher/scientist at Microsoft AI & Research, Washington and GE Global Research, California. I got my B.S. (Math) from Fudan University and Ph.D. from the University of Maryland, Baltimore County.


My research topics primarily covers machine (deep) learning, non-convex optimization

and their frontier applications. My previous work covers the state-of-the-art research

and large-scale landings in


  • NLP/Conversational AI

  • Medical Image Analysis/Computer Vision

  • Ads, Search and User Understanding

  • Digital Signal Modeling (time series, acoustics)

  • Deep Reinforcement Learning


I am also serving as the reviewer/PC for top conferences like IJCAI, AAAI, KDD, NIPS, CVPR, WWW, WSDM, ACL, EMNLP, NAACL and journals like PLoS One, TNNLS, Neural Networks, Pattern Recognition, NeuroComputing, etc


Technical Report

Publication

1. Diabetic Retinopathy Detection via Deep Convolutional Networks for Discriminative Localization and Visual Explanation. [PDF][code]

Zhiguang Wang and Jianbo Yang

Proceedings of The Thirtieth AAAI Conference on Artificial Intelligence, 2018. (AAAI)

2. Encoding Temporal Markov Dynamics in Graph for Time Series Visualization. [PDF]

Liu, Lu, and Zhiguang Wang.

Proceedings of The Thirtieth AAAI Conference on Artificial Intelligence, 2018. (AAAI)

3. Time Series Classification from Scratch with Deep Neural Networks: A Strong Baseline. [PDF][Code]

Zhiguang Wang, Weizhong Yan, Tim Oates

Proceedings of The International Joint Conference on Neural Networks, 2017. (IJCNN)

4. Empirical study of symbolic aggregate approximation for time series classification.

Song, Wei, Zhiguang Wang, Fan Zhang, Yangdong Ye, and Ming Fan.

Journal of Intelligent Data Analysis 21, no. 1 (2017): 135-150. (IDA)

5. Deep learning for unsupervised feature extraction in audio signals: Monaural source separation.

Edward T Nykaza, Arnold P Boedihardjo, Zhiguang Wang, Tim Oates, Anton Netchaev, Steven L Bunkley, Matthew G Blevins

The Journal of the Acoustical Society of America 140(4):3424-3424 · October 2016

6. Deep learning for unsupervised feature extraction in audio signals: A pedagogical approach to understanding how hidden layers recreate, separate, and classify audio signals.

Edward T Nykaza, Arnold P Boedihardjo, Zhiguang Wang, Tim Oates, Anton Netchaev, Steven L Bunkley, Matthew G Blevins

The Journal of the Acoustical Society of America 139 (4), 2069-2069, 2016. (ASA)

7. Adaptive Normalized Risk-Averting Training For Deep Neural Networks. [PDF][Code]

Z Wang, T Oates, J Lo

Proceedings of The Thirtieth AAAI Conference on Artificial Intelligence, 2016. (AAAI)

8. Imaging Time-Series to Improve Classification and Imputation. [PDF] [Code]

Z Wang, T Oates

Proceedings of the 24th International Conference on Artificial Intelligence, Pages 3939-3945, 2015. (IJCAI)

9. Pooling SAX-BoP Approaches with Boosting to Classify Multivariate Synchronous Physiological Time Series Data. [PDF]

Z Wang, T Oates

Proceedings of the Twenty-Eighth International Florida Artificial Intelligence Research Society Conference, 335-341, 2015. (FLAIRS)

10. Encoding Time Series as Images for Visual Inspection and Classification Using Tiled Convolutional Neural Networks. [PDF]

Z Wang, T Oates

Workshops at the Twenty-Ninth AAAI Conference on Artificial Intelligence, 2015.

11. Empirical Studies on Symbolic Aggregation Approximation Under Statistical Perspectives for Knowledge Discovery in Time Series. [PDF]

Zhiguang Wang, Wei Song, Yangdong Ye, Ming Fan

Proceedings of the 12th International Conference on Fuzzy Systems and Knowledge Discovery, 2015.

12. Time warping symbolic aggregation approximation with bag-of-patterns representation for time series classification. [PDF]

Z Wang, T Oates

Proceedings of the 13th International Conference on Machine Learning and Applications, 2014.

13. Control channels in the brain and their influence on brain executive functions.

Qinglei Meng, Fow-Sen Choa, Elliot Hong, Zhiguang Wang, Mohammad Islam

Proc. SPIE 9107, Smart Biomedical and Physiological Sensor Technology XI, 910716, 2014.

Service

  • PC member: IJCAI, AAAI, KDD, ACL, NAACL, EMNLP, KDD, WSDM, WWW, NIPS, CVPR

  • Reviewer: IEEE Transactions on Neural Networks and Learning Systems, PLOS ONE, IEEE Transactions on Cybernetics, IEEE Access, NeuroComputing, Pattern Recognition, Neural Networks, Financial Innovations, Statistical Methods in Medical Research


Link
LinkedIn