Ying Xiao (肖盈)
I am currently a Senior Staff Research Scientist at Twitter Cortex. Previously, I was at Google Research for 4.5 years, Palantir Technologies, and Goldman Sachs. I group/uber tech lead half a dozen recommendation systems projects at Twitter, including both research and production projects; additionally, I directly manage a small team of research scientists. My work is primarily around speed, scale, and architectures for heavy ranker models in recommendation systems.
I graduated with a PhD in machine learning theory from Georgia Tech where I was advised by Santosh Vempala.
Email: ying.[chemical symbol for iron].xiao@gmail.com
Research
I have unusally broad research interests in machine learning. Currently, I work on recommender systems, but I bring a broad perspective and deep experience from my previous work. Previously, I have worked on text-to-speech (I was a member of the Tacotron team, building one of the first successful end-to-end tts models), deep learning optimization (developed a technique to understand Hessian eigenspectra of real DNNs), optical character recognition, and machine learning theory.
Ahmed El-Kishky, Thomas Markovich, Kenny Leung, Frank Portman, Aria Haghighi, Ying Xiao, kNN-Embed: Locally Smoothed Embedding Mixtures for Multi-interest Candidate Generation, (submitted 2022).
Ahmed El-Kishky, Thomas Markovich, Serim Park, Chetan Verma, Baekjin Kim, Ramy Eskander, Yury Malkov, Frank Portman, Sofía Samaniego, Ying Xiao, Aria Haghighi, TwHIN: Embedding the Twitter Heterogeneous Information Network for Personalized Recommendation, (submitted 2022).
Behrooz Ghorbani, Shankar Krishnan, Ying Xiao, An investigation into neural net optimization via hessian eigenvalue density, ICML 2019.
Siyang Qin, Alessandro Bissacco, Michalis Raptis, Yasuhisa Fujii, Ying Xiao, Towards Unconstrained End-to-End Text Spotting, ICCV 2019.
Shankar Krishnan, Ying Xiao, Rif A Saurous, Neumann Optimizer: A Practical Optimizer for Deep Neural Networks, ICLR 2018.
RJ Skerry-Ryan, Eric Battenberg, Ying Xiao, Yuxuan Wang, Daisy Stanton, Joel Shor, Ron Weiss, Rob Clark, Rif A Saurous, Towards end-to-end prosody transfer for expressive speech synthesis with tacotron, ICML 2018.
Yuxuan Wang, Daisy Stanton, Yu Zhang, RJ-Skerry Ryan, Eric Battenberg, Joel Shor, Ying Xiao, Ye Jia, Fei Ren, Rif A Saurous, Style tokens: Unsupervised style modeling, control and transfer in end-to-end speech synthesis, ICML 2018.
Yuxuan Wang, RJ Skerry-Ryan, Daisy Stanton, Yonghui Wu, Ron J Weiss, Navdeep Jaitly, Zongheng Yang, Ying Xiao, Zhifeng Chen, Samy Bengio, Quoc Le, Yannis Agiomyrgiannakis, Rob Clark, Rif A Saurous, Tacotron: towards end-to-end speech synthesis, Interspeech 2017.
Raffay Hamid, Ying Xiao, Alex Gittens and Dennis Decoste, Compact Random Feature Maps, ICML 2014.
Author lists are ordered lexicographically in theoretical computer science:
Santosh S Vempala, Ying Xiao, Max vs min: Tensor decomposition and ICA with nearly linear sample complexity, COLT 2015.
Navin Goyal, Santosh Vempala and Ying Xiao, Fourier PCA , STOC 2014.
Vitaly Feldman, Elena Grigorescu, Lev Reyzin, Santosh Vempala and Ying Xiao, Statistical Algorithms and a Lower Bound for Planted Clique, STOC 2013.
Santosh Vempala and Ying Xiao, Complexity of Learning Subspace Juntas and ICA , Asilomar Conference on Signals, Systems & Computers 2013.
Contact Information
ying.[chemical symbol for iron].xiao@gmail.com