Hi, Welcome to my homepage! I started my Ph.D in Fall 2006 with Artificial Intelligence Group and my boss is Prof. Prasad Tadepalli :-) . My general research interests are machine learning and data mining. Current focus of my research is Structured Prediction models and Integrated learning. I also collaborate with Prof. Alan Fern on learning to search, and Prof. Lise Getoor on social network analysis problems. Generally, I work on anything that is fun and sufficiently challenging!
I also organize the "learning to search" and "machine reading" reading groups.
NEW: Our team "Chak De Oregon" won the NWCL cricket championship this year! I maintain my cricket scores here. My friends think I'm crazy to travel to Washington every week for playing cricket :-) I'm hoping to play in Canadian league next season based on my schedule!!!
M.Tech, Indian Institute of Technology, Kanpur, July 2006.
B.Tech, Andhra University College of Engineering, Visakhapatnam, April 2004.
Janardhan Rao Doppa, Jun Yu, Lise Getoor and Prasad Tadepalli: Chance-Constrained Programs for Link Prediction , to appear in Workshop on Analyzing Networks and Learning with Graphs at NIPS 2009.
Zhang, Yoon, Dibona, Appling, Ding, Doppa, (and other IL team members) An Ensemble Learning and Problem Solving Architecture for Airspace Management , Proceedings of IAAI/IJCAI 2009 (PDF) .
Machine Reading (with Prof. Prasad Tadepalli, Prof. Tom Dietterich, Prof. Xiaoli Fern, Mohammad and David Weiser)
We are working on this DARPA project named "Machine Reading" along with Tom Mitchell's group from CMU, Oren Etzioni's group from Univ. of Washington, ISI and BBN technologies. Our OSU team is working on learning inference rules from the predicate graphs, to help the inference engine in answering questions.
Integrated Learning (with Prof. Prasad Tadepalli)
How can we integrate information from multiple sources to learn better ? We are trying to understand the tradeoff between multiple learners combining their predictions vs a single learner which uses all the views of the data, from the PAC learning thoery point of view. We consider both multi-class and structured prediction tasks for this problem. In the past, we worked on DARPA's Integrated learning project (GILA), where the goal was to learn a complex problem solving task from a single demonstration of the expert. Our part was to learn the utility or cost function of the expert he/she is minimizing subject to some constraints. We formulate it as a constrained optimization problem and solve it using structured gradient boosting. We call this DTLR(Decision Theoretic Learner and Reasoner). We worked with other researchers from Lockheed-Martin, ASU, RPI, UMD, UMASS and Georgia Tech.
Learning Search Control for Complex prediction problems (with Prof. Prasad Tadepalli and Prof. Alan Fern)
How can we learn to control search for complex prediction problems like structured prediction, planning and reinforcement learning ?. We are exploring the space of algorithms that integrate search into learning, to study their behavior w.r.t the above mentioned problems.
What's next ?:
For the later half of my Ph.D, I would like to work on another open problem - "How to integrate reasoning into learning for structured prediction models?". I would also like to apply structured prediction models to various other problems in Computer Vision and Biology.
Student volunteer for ICML/ILP 2007
Reviewer :
ICML 2008, IJCAI 2009
Machine Learning Journal
Artificial Intelligence, Machine Learning, Bayesian Networks, Planning and Reinforcement Learning, First-order logic and Relational learning, Computer Vision, Special Topics in High-level Computer Vision
Theory of Statistics-1, Theory of Statistics-2, Non-linear Optimization, Linear Algebra
Error-Correcting Codes, Algorithms, Analysis of Algorithms, Theory of Computation, Advanced data structures and algorithms, Advanced graph algorithms
ST/HCI meets software engineering, Programming languages
I was Instructor for the following courses:
C101: Unix and C Programming, offered by programming club, IIT Kanpur (single class of almost 160 students!!) Fall 2005
Summer School on Data structures and Algorithms, offered by Dept. of Computer Science, IIT Kanpur. (Summer 2006)
Object Oriented Programming in C++, offered by programming club, IIT Kanpur. (Winter 2006)
CS261 Data Structures, Oregon State University. (Summer 2007)
I was Teaching Assistant for the following courses:
Esc101: Java Programming Lab, IIT Kanpur, Fall 2004.
CS365: Artificial Intelligence Programming, IIT Kanpur, Dr. Pabitra Mitra (with Barna Saha)
CS430: Introduction to Artificial Intelligence, Oregon State University, Dr. Weng-Keen Wong
CS271: Computer Architecture and Assembly Language Programming, Oregon State University, Dr. Paulson
CS411: Operating Systems-II (Linux Kernel Development), Oregon State University, Dr. Paulson (with Rob, Nick and Yun-Rim)
Scholarship from the Ministry of Human Resource and Development, India (2004-06), as a GATE (Graduate Aptitude Test for Engineers)-qualified student with 99.8 percentile and AIR (All India Rank) 75.
Scored 99 percent in Maths paper for the Engineering and Medical entrance examination (2000).
Sir. C.V. Raman Educational Award (1998), awarded by the State Govt. of Andhra Pradesh, India.