What's New

The webpage on kernelized LSH is online here with new results and code

"Fast Similarity Search for Learned Metrics" was chosen as the featured article for the Dec. 2009 issue of IEEE Transactions on Pattern Analysis and Machine Intelligence

"Learning to Hash with Binary Reconstructive Embeddings" was accepted at NIPS as a spotlight presentation

Introduction

I am a postdoc at UC Berkeley EECS and ICSI, hosted by Trevor Darrell. I finished my Ph.D. in computer science in November, 2008, supervised by Inderjit Dhillon in the University of Texas at Austin computer science department. I did my undergrad in computer science and mathematics at Cornell University.

Research Interests:

• Machine Learning
• Data Mining / Data Analysis
• Numerical Optimization
• Applications to Computer Vision

The goal of my research is to make it easier to analyze and search complex, large-scale data. A major focus of my research is on large-scale optimization for core problems in machine learning such as metric learning, content-based search, and large-scale graph clustering. Furthermore, I am interested in large-scale applications of machine learning to practical problems, especially in computer vision. I have worked with John Platt and Arun Surendran at Microsoft Research on large-scale optimization, and as an undergraduate, I worked with John Hopcroft on tracking topics in networked data over time. During the Fall 2007 semester, I was a research fellow at the Institute for Pure and Applied Mathematics at U.C.L.A.

Curriculum Vitae [pdf]

Research Details

Click here to read more about some of my research.

Publications by Type

Publications Chronologically  

Contact Info

Email: kulis [at] eecs [dot] berkeley [dot] edu