University of Southern California
Research Group
Amulya Yadav

Update: I am joining Penn State University as an Assistant Professor this Fall. My new website is here.

Ph.D. Completed! Passed my thesis defense.

I am a final year Ph.D. candidate in the Computer Science Department at the University of Southern California (USC), and I work in the Teamcore Research Group, while being advised by Prof. Milind Tambe. I plan to graduate by April 2018.

This is a brief statement of my current and future research: Research Statement

Before coming to USC, I worked as a Software Development Engineer at and even before that, I spent four amazing years at Indian Institute of Technology Patna as a CS undergraduate student.

Recent Honors:
1. Won Best Research Assistant Award at USC for the academic year 2016-2017.

2. Won Best Paper Nomination for AAMAS 2017 paper (7 out of 288 papers).

3. CCC Great Innovative Idea for May 2017 .

4. Won Best Video & Best Student Video Award at AAAI 2017.

5. Won Pragnesh Jay Modi Best Student Paper Award for AAMAS 2016 paper.

6. Won Most Visionary Paper Award for IDEAS 2016 paper.

7. My work was highlighted by as one of 26 incredible innovations that improved the world in 2015.

Recent News:
1. Finished internship at Microsoft Research India

2. PC Member of IJCAI 2017

3. Passed my thesis proposal.

5. Article published in the AI Magazine.


Broadly speaking, I am most interested in solving real-world problems using techniques from Game Theory and AI. Currently, I have been looking at ways to reduce incidence of HIV spread amongst homeless populations using techniques from AI planning. More details about this work can be found here.

Right now, I am also trying to get this work deployed in the real world by collaborating with My Friend's Place, a homeless shelter which provides relief(food, clothing, shelter and rehabilitation) to homeless youth in Los Angeles.

Here is some recent media coverage about this project which can help to give you a high level idea about the project and the research problems that arise in this domain:

Other domains which I would like to explore in the near future include:

  • Negotiation techniques in climate change conventions to attain optimal social welfare. My motivation for pursuing this is my frustration at the inability of world nations to reach meaningful conclusions on climate change, despite the urgent need to do so. More specifically, is there some non-trivial protocol for negotiation which nations need to made aware of before they enter these climate change conventions?

  • Stock market prediction using game theory. Game theory applies very naturally to this domain. However, as far as I know, people usually rely on machine learning techniques to make such predictions.

  • Can game theory be used to help make more well informed foreign policy decisions? I am interested in exploring if there is a game-theoretic explanation for why conflicts(India-Pakistan, Israel-Palestine etc.) continue to remain unsolved. I believe that doing so would help me differentiate more clearly between game theory on paper and game theory in the real world.

I am also interested in learning more about machine learning applications in industry. One particular thing that has piqued my interest recently is the functioning of ad exchanges.

If you are interested in my work, feel free to contact me: 
LinkedIn Profile: Link
Snail mail: SAL 300, 941 Bloom Walk, Los Angeles, CA 90089