Tianshi Gao's Homepage

Bio

I'm a research scientist at Facebook working on practical large-scale machine learning systems. I graduated from Stanford University with a Ph.D. in Electrical Engineering in 2012. Before that, I obtained my bachelor's degree in EE with honors from Tsinghua University in 2007.

My Ph.D. advisor at Stanford is Professor Daphne Koller who has co-founded Coursera .


Publications

What Makes a Good Detector? -- Structured Priors for Learning From Few Examples
Tianshi Gao, Michael Stark and Daphne Koller
In Proceedings of European Conference on Computer Vision (ECCV), 2012.
[pdf | bibtex]

Active Classification Based on Value of Classifier
Tianshi Gao and Daphne Koller
In Advances in Neural Information Processing Systems (NIPS), 2011. (Spotlight presentation)
[pdf | bibtex]

Discriminative Learning of Relaxed Hierarchy for Large-scale Visual Recognition
Tianshi Gao and Daphne Koller
In IEEE International Conference on Computer Vision (ICCV), 2011. (Oral presentation)
Also presented at First Workshop on Fine-Grained Visual Categorization in conjunction with CVPR 2011 .
(Best Poster Runner-up Award)
[pdf | supplementary material | bibtex | code (v0.1)]

Multiclass Boosting with Hinge Loss based on Output Coding
Tianshi Gao and Daphne Koller
In Proceedings of the International Conference on Machine Learning (ICML), 2011.
[pdf | bibtex]

A Segmentation-aware Object Detection Model with Occlusion Handling
Tianshi Gao, Benjamin Packer and Daphne Koller
In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011.
[pdf | supplementary material | bibtex]

Region-based Segmentation and Object Detection
Stephen Gould, Tianshi Gao and Daphne Koller
In Advances in Neural Information Processing Systems (NIPS), 2009. (Spotlight presentation)
[pdf | bibtex]

some old work

Filed Patents

US 61/176,815, Storing and Searching Fingerprints Derived From Media Content Based On a classification of the Media Content,
with Regunathan Radhakrishnan, Wenyu Jiang and Claus Bauer in the research department of Dolby Laboratories.

US 61/186,353, Trend Analysis In Content Identification Based On Fingerprinting,
with Wenyu Jiang in the research department of Dolby Laboratories.


Teaching

Teaching Assistant, CS228: Structured Probabilistic Models: Principles and Techniques, Winter 2009 and Winter 2010.

Teaching Assistant, CS294A: Projects in Holistic Scene Understanding, Autumn 2009 and Autumn 2010.


Contact

Address:
1601 Willow Road
Menlo Park, CA 94025

Email:
picture