I am an assistant professor in the department of Computer Science at University of Virginia. My research goal is to build intelligence systems that solve real-world problems by automatically acquiring knowledge. This challenging goal involves two fundamental components: A machine learning component that can efficiently make coherent decisions for problems with complex structures, and a natural language understanding component that enables the system to extract knowledge from unstructured text. I have been published broadly in machine learning, natural language processing, artificial intelligence, and data mining.
If you are interested in joining my group, please read my Research Statement written for prospective students. However, due to the large number of these inquiries, I generally cannot respond personally. If you're a UVA student, please drop by my office hour.
- email: kw AT kwchang DOT net
- office: R412, Rice Hall, University of Virginia
- office hour: 2:00pm --3:00pm Tue.
- see my calendar for scheduling a meeting with me.
- 03.11.2017, My proposal, CRII: Learning Structured Prediction Models with Auxiliary Supervision has been funded by NSF.
- 01.25.2017, Congratulate Wasi Ahmad on being awarded the William L Ballard Jr. Fellowship for Spring of 2017.
- 12.06.2016, I gave an invited talk at NIPS 16 Workshop on Learning in High Dimensions with Structure and VW tutorial at Machine Learning System Workshop
- 11.28.2016, Our paper is accepted by AAAI 17: Structured Prediction with Test-time Budget Constraints
- 11.08.2016, Thanks all participants for a successful workshop. See you in the Second Workshop on Structured Prediction for NLP at EMNLP 2017 in Copenhagen!
- 08.12.2016, Our paper about debiasing wordembedings is reported by NPR and MIT Technology Review.
- 08.12.2016, Two papers accepted by NIPS! Credit assignment compiler for L2S; Debiasing wordembeddings.
- 08.10.2016, I join the Computer Science Department at University of Virginia as an Assistant Professor!
- 08.01.2016, One paper accepted by EMNLP: Learning from Explicit and Implicit Supervision Jointly For Algebra Word Problems
- 02.13.2016, Checkout our tutorial on Structured Prediction at AAAI16.
- Spring 2017, CS6501: Advanced Machine Learning -- Structured Prediction and Deep Learning, University of Virginia
- Fall 2016, CS6501: Natural Language Processing, University of Virginia
Softwares & Demos
- Illinois-SL: Package for learning structured prediction models.
- Liblinear: A library for large-scale linear classification.
- Vowpal Wabbit: An online learning system (I contributed to several L2S applications)
- Coreference Resolution Demo: Identify noun phrases that refer to the same entity.
- MSR Continuous Space Text Representations: Measure the degree of relation of two words. (I contributed to MRLSA).
- Package for learning with limited memory: Learning a linear classifier when data cannot fit in memory.
Experience and Education (Curriculum Vitae)
- Post-doc, Microsoft Research New England 2015-2016
- Ph.D., Department of Computer Science, University of Illinois at Urbana-Champaign 2010--2015
- M.S. (Computer Science), National Taiwan University, 2007--2009
- B.S. (Computer Science), B.S. (Electrical Engineering), National Taiwan University, 2003--2007
- Summer intern, Microsoft Research New York (2014), Microsoft Research Redmond (ML group, 2013), Microsoft Cloud and Information Services Lab (2012), Google Taiwan (2008)