Agent Management

Chongjie Zhang

Department of Computer Science Phone: 413-***-****

University of Massachusetts Amherst Fax: 413-***-****

*** ********* ***** *****: ********@**.*****.***

Amherst, MA 01003 Web: http://www.cs.umass.edu/ chongjie

~

RESEARCH INTERESTS

Artificial intelligence, multi-agent systems, multi-agent learning, reasoning under

uncertainty, dis-

tributed optimization, resource allocation, game theory, and machine learning

EDUCATION

University of Massachusetts Amherst, Amherst, MA, USA (2006 { 2011)

Ph.D. in Computer Science

Advisor: Professor Victor Lesser

Thesis: Scaling Multi-Agent Learning in Complex Environments

Committee: Victor Lesser (chair), Andrew Barto, Shlomo Zilberstein, and Jonathan Machta

Louisiana State University, Baton Rouge, LA, USA (2004 { 2006)

Master in Systems Science

Advisor: Professor Gabrielle Allen

University of International Business and Economics, Beijing, China (1997 { 2001)

Bachelor in Economics

RESEARCH & WORK EXPERIENCE

Post-Doctoral Research Associate 2011 { Present

Advisor: Prof. Victor Lesser University of Massachusetts Amherst

Research areas: multi-agent learning, planning under uncertainty, distributed

optimization, game

theory, agent organization, multi-agent systems, artificial intelligence

Research Assistant 2006 { 2011

Advisor: Prof. Victor Lesser University of Massachusetts Amherst

Research areas: multi-agent learning, planning under uncertainty, distributed

optimization, game

theory, agent organization, multi-agent systems, artificial intelligence

Teaching Assistant 2009 Fall

Instructor: Prof. William Verts University of Massachusetts Amherst

Course: CMPSCI 105: Computer Literacy

Research Assistant2004 { 2006

Advisor: Prof. Gabrielle Allen Louisiana State University

Research areas: grid computing, machine learning, and Internet technology

Software Engineer2001 { 2004Hansky Ltd. Beijing, China

Working areas: software configuration management, software requirement management

Chief Architectof a software requirement management product (Hansky Dragonfly)

Technical Leaderof a change request management product (Hansky Butterfly)

Developerof a software asset management product (Hansky Frirefly)

RESEARCH PROJECTS

Coordinated Multi-Agent Reinforcement Learning

The goal of this project is to design e cient learning approaches that allow agents to

adapt

to the dynamics of the environment and other agents. We seek approaches that are both

theoretically-founded and perform well in practical applications. We developed a new

multi-

agent learning algorithm with formal convergence analysis in games. To scale multi-agent

learning algorithms to large applications, we developed two approaches to coordinating

multi-

agent learning. The first approach introduces amulti-level supervisory control

frameworkthat

exploits non-local information to coordinate agents' learning processes. The second

approach

synthesizes multi-agent reinforcement learning and distributed constraint optimization to

ex-

ploit structured interaction among agents. This approach provably learns globally optimal

policies in a distributed, scalable manner for a class of cooperative problems.

(2006 { Present)

Distributed Resource Allocation

We developed different approaches for two resource allocation domains: 1)surveillance

systems

and 2)cloud computing. In collaboration with a research team from Raytheon, we

investigated

multi-level negotiation strategies for optimizing online resource allocation for

surveillance sys-

tems that contain resources privately owned by multiple different organizations. For

cloud

computing, we developed a multi-agent learning approach to dynamically improving polices

for distributed resource managers that collaboratively allocate resources for tasks.

(2008 { Present)

Designing Organizationally Adept Agent

The centerpiece of this project is the design, development, and evaluation of

computational

representations and algorithms for making software agents organizationally adept in large-

scale multi-agent organizations. We developed an extended BDI reasoning architecture for

an

organizationally adept agent that balances organizational, social, and agent-centric

interests

to autonomously adapt to achieve organizational objectives emergently.

(2010 { Present)

Integrated Learning (GILA)

We developed approaches to coordinating multiple learning agents, each with a different

learn-

ing algorithm, and integrating their hypotheses to cooperatively and incrementally solve

com-

plex problems. We investigated learning algorithms that can improve such coordination and

cooperation. Air tra c flight planning was the primary application of this research

project.

(2006 { 2008)

Collaborative Adaptive Sensing of the Atmosphere (CASA)

This project developed distributed sensing networks that are composed of radars which op-

erate collaboratively within a dynamic information technology infrastructure and adapt to

changing atmospheric conditions. We designed scheduling algorithms to optimize radar

scans

for detecting atmospheric hazards.

(2007)

Ubiquitous Computing and Monitoring System (UCoMS)

This research aims to develop and deploy a Ubiquitous Computing and Monitoring System

(UCoMS) for discovery and management of energy resources. We built response surface

models

for oil reservoir simulations by using Radial Basis Function Networks and explored

sampling

techniques to collect data e ciently.

(2005 { 2006)

SURA Coastal Ocean Observing and Prediction (SCOOP)

We designed and built a grid portal for the coastal research community to reduce the

complexity

of grid computing for end users and enable advanced application scenarios. The resulting

portal

uses new collaborative tools to better access ocean data, and computational resources.

(2004 { 2006)

HIGHLY REFERENCED CONFERENCE PUBLICATIONS

Chongjie Zhang and Victor Lesser. Coordinating Multi-Agent Reinforcement Learning in

Networked

Distributed POMDPs. In Proceedings of the 25th National Conference on Artificial

Intelligence

(AAAI), 2011.

Chongjie Zhang and Victor Lesser. Multi-Agent Learning with Policy Prediction.

InProceedings of

REFERENCED WORKSHOP AND OTHER PUBLICATIONS

Daniel D. Corkill, Edmund H. Durfee, Victor R. Lesser, Huzaifa Zafar, and Chongjie Zhang.

Or-

ganizationally Adept Agents. In Proceedings of the 12th International Workshop on

Coordination,

Organizations, Institutions and Norms in Agent Systems (COIN), 2011.

Gabrielle Allen, Promita Chakraborty, Dayong Huang, Zhou Lei, John Lewis, Christopher

White,

Xiaoxi Xu, and Chongjie Zhang. A Workflow Approach to Designed Reservoir Study.

InProceedings

of the 2nd Workshop on Workflows in Support of Large-Scale Science, WORKS '07. ACM Press,

New York, NY, 75-79, June 2007.

Jon MacLaren, Gabrielle Allen, Chirag Dekate, Dayong Huang, Andrei Hutanu, and Chongjie

Zhang.

Shelter from the Storm: Building a Safe Archive in a Hostile World. InLecture Notes in

Computer

Science, Volume 3762, p. 294, 2005.

Chongjie Zhang, Chirag Dekate, Gabrielle Allen, Ian Kelley, and Jon MacLaren. An

Application

Portal for Collaborative Coastal Modeling. In thefirst International Workshop on Grid

Computing

Environments, in conjunction with Supercomputing Conference, Seattle, WA, 2005 (Best

Paper

Award)

Chongjie Zhang, Ian Kelley, and Gabrielle Allen. Grid Portal Solutions: A Comparison of

Grid-

Portlets and OGCE. In the first International Workshop on Grid Computing Environments, in

conjunction with Supercomputing Conference, Seattle, WA, 2005

HONORS & AWARDS

Phi Kappa Phi Honor Society: Elected Member

Best Paper Award, \An Application Portal for Collaborative Coastal Modeling", the first

In-

ternational Workshop on Grid Computing Environments, in conjunction with Supercomputing

Conference, Seattle, WA, 2005

Yearly Merit Award, Hansky Ltd, Beijing, China, 2003

Academic Excellence Fellowship, University of International Business and Economics,

Beijing,

China, 1999 { 2000Sumitomo Bank Scholarship for Excellent Students, University of International Business

and

Economics, Beijing, China, 1998

PROFESSIONAL SERVICES

Program Committee:

MSDM-2012: Seventh Workshop in Multiagent Sequential Decision Making Under Uncertainty

Reviewing:

Computational Intelligence

IJCAI-11: Twentieth International Joint Conference of Artificial Intelligence

AAAI-12: Twenty-Sixth Conference on Artificial Intelligence

Contact this candidate