Automated Assistants for Analyzing Team Behaviors

Multi-agent teamwork is critical in a large number of agent applications, including training, education, virtual enterprises and collective robotics. The complex interactions of agents in a team as well as with other agents make it extremely difficult for human developers to understand and analyze agent-team behavior. It has thus become increasingly important to develop tools that can help humans analyze, evaluate, and understand team behaviors. However, the problem of automated team analysis is largely unaddressed in previous work. In this article, we identify several key constraints faced by team analysts. Most fundamentally, multiple types of models of team behavior are necessary to analyze different granularities of team events, including agent actions, interactions, and global performance. In addition, effective ways of presenting the analysis to humans is critical and the presentation techniques depend on the model being presented. Finally, analysis should be independent of underlying team architecture and implementation.We also demonstrate an approach to addressing these constraints by building an automated team analyst called ISAAC for post-hoc, off-line agent-team analysis. ISAAC acquires multiple, heterogeneous team models via machine learning over teams' external behavior traces, where the specific learning techniques are tailored to the particular model learned. Additionally, ISAAC employs multiple presentation techniques that can aid human understanding of the analyses. ISAAC also provides feedback on team improvement in two novel ways: (i) It supports principled “what-if” reasoning about possible agent improvements; (ii) It allows the user to compare different teams based on their patterns of interactions. This paper presents ISAAC's general conceptual framework, motivating its design, as well as its concrete application in two domains: (i) RoboCup Soccer; (ii) software agent teams participating in a simulated evacuation scenario. In the RoboCup domain, ISAAC was used prior to and during the RoboCup '99 tournament, and was awarded the RoboCup Scientific Challenge Award. In the evacuation domain, ISAAC was used to analyze patterns of message exchanges among software agents, illustrating the generality of ISAAC's techniques. We present detailed algorithms and experimental results from ISAAC's application.

[1]  John Beidler,et al.  Data Structures and Algorithms , 1996, Wiley Encyclopedia of Computer Science and Engineering.

[2]  Ehud Reiter,et al.  Has a Consensus NL Generation Architecture Appeared, and is it Psycholinguistically Plausible? , 1994, INLG.

[3]  Nicholas R. Jennings,et al.  Controlling Cooperative Problem Solving in Industrial Multi-Agent Systems Using Joint Intentions , 1995, Artif. Intell..

[4]  Phoebe Sengers,et al.  Designing Comprehensible Agents , 1999, IJCAI.

[5]  Milind Tambe,et al.  Execution monitoring in multi-agent environments , 2000 .

[6]  Thomas Rist,et al.  A standard reference model for intelligent multimedia presentation systems , 1997, Comput. Stand. Interfaces.

[7]  Tomoichi Takahashi.  LogMonitor: From Player's Action Analysis to Collaboration Analysis and Advice on Formation , 1999, RoboCup.

[8]  Hiroaki Kitano,et al.  The RoboCup Synthetic Agent Challenge 97 , 1997, IJCAI.

[9]  Katia P. Sycara,et al.  Distributed Intelligent Agents , 1996, IEEE Expert.

[10]  Young,et al.  Inferring statistical complexity. , 1989, Physical review letters.

[11]  Alberto Maria Segre,et al.  Programs for Machine Learning , 1994 .

[12]  Wei-Min Shen,et al.  Dynamic Distributed Resource Allocation: A Distributed Constraint Satisfaction Approach , 2001, ATAL.

[13]  Manuela M. Veloso,et al.  Using decision tree confidence factors for multi-agent control , 1998, AGENTS '98.

[14]  Hyacinth S. Nwana,et al.  Visualization and Debugging of Distributed Multiagent Systems , 1999, Appl. Artif. Intell..

[15]  Milind Tambe,et al.  Automated assistants to aid humans in understanding team behaviors , 2000, AGENTS '00.

[16]  Luc De Raedt,et al.  Inductive Verification and Validation of the KULRoT RoboCup Team , 1998, RoboCup.

[17]  Milind Tambe,et al.  The Benefits of Arguing in a Team , 1999, AI Mag..

[18]  Kai Ming Ting,et al.  Inducing Cost-Sensitive Trees via Instance Weighting , 1998, PKDD.

[19]  Milind Tambe,et al.  Towards Flexible Teamwork , 1997, J. Artif. Intell. Res..

[20]  Milind Tambe,et al.  Building Dynamic Agent Organizations in Cyberspace , 2000, IEEE Internet Comput..

[21]  Maja J. Mataric,et al.  Coordinating mobile robot group behavior using a model of interaction dynamics , 1999, AGENTS '99.

[22]  Tucker R. Balch.  The impact of diversity on performance in multi-robot foraging , 1999, AGENTS '99.

[23]  北野 宏明,et al.  RoboCup-97 : robot soccer World Cup I , 1998 .

[24]  S. Toulmin.  The uses of argument , 1960 .

[25]  W. Lewis Johnson,et al.  Agents that Learn to Explain Themselves , 1994, AAAI.

[26]  Geoffrey I. Webb,et al.  Using Decision Trees for Agent Modeling: Improving Prediction Performance , 2004, User Modeling and User-Adapted Interaction.

[27]  I WebbGeoffrey,et al.  Using Decision Trees for Agent Modeling , 1998 .

[28]  Sholom M. Weiss,et al.  Using Empirical Analysis to Refine Expert System Knowledge Bases , 1984, Artif. Intell..

[29]  Rich Caruana,et al.  Greedy Attribute Selection , 1994, ICML.

[30]  Dana Ron,et al.  On the learnability and usage of acyclic probabilistic finite automata , 1995, COLT '95.

[31]  Milind Tambe,et al.  Intelligent Agents for Interactive Simulation Environments , 1995, AI Mag..

[32]  Milind Tambe,et al.  Agent Assistants for Team Analysis , 2000, AI Mag..

[33]  Andreas Stolcke,et al.  Hidden Markov Model} Induction by Bayesian Model Merging , 1992, NIPS.

[34]  Kôiti Hasida,et al.  MIKE: an automatic commentary system for soccer , 1998, Proceedings International Conference on Multi Agent Systems (Cat. No.98EX160).

[35]  Inderpal S. Bhandari,et al.  Advanced Scout: Data Mining and Knowledge Discovery in NBA Data , 2004, Data Mining and Knowledge Discovery.

[36]  Stacy Marsella,et al.  An Instructor's Assistant for Team-Training in Dynamic Mulit-Agent Vitual Worlds , 1998, Intelligent Tutoring Systems.

[37]  Debra Schreckenghost,et al.  Adjustable Autonomy for Human-Centered Autonomous Systems , 1999 .

[38]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[39]  Ian Frank,et al.  Soccer Server: A Tool for Research on Multiagent Systems , 1998, Appl. Artif. Intell..