CSCI 599: AI for Social Good

Location: KAP 167

Time: Monday 3 to 620 PM

 

Instructors: Milind Tambe/Eric Rice

 

Introduction

AI for Social Good is a course which is being jointly taught by Dr. Milind Tambe (Viterbi School of Engineering) and Dr. Eric Rice (Suzanne Dworak-Peck School of Social Work).  The purpose of the class is to expose doctoral students and advanced masters level students to a new field of research which merges research techniques in artificial intelligence with theory and research contexts from social work.  The course is attached to Drs. Tambe and Rice’s new Center for Artificial Intelligence in Society (CAIS).  This course is intended to be trans-disciplinary.  It will be a small seminar style course, with limited enrollment. The intention is to have half of the student body from engineering and social work respectively. There is no need for programming experience or social work practice experience per se, however, a deep connection to either social work or computer science is needed.  The course will provide an overview of major methods that can advance this new trans-disciplinary work, from both a conceptual level in computer science and a conceptual level in social work.  Topics from engineering will include decision theory, sequential planning under uncertainty, and machine learning predictive models.  Topics from social work will include human behavior theories that can be modeled by AI (such as social cognitive theory and social network theory) as well as context driven topics, such as homelessness and health care access.  The course meets weekly, is a discussion style course and will include an emphasis on group projects which are tackled in trans-disciplinary teams. 

PhD students are expected to participate heavily in teaching concepts to the other half of the class. PhD students should team up in exercises. Each class will typically introduce concepts and generate some exercises for the class to be jointly done.

 

 

 

 

 

 

Syllabus

Jan 9:

Class mechanics

Rice: Introduction to Social work

Readings:

Brekke, J. S. (2012). Shaping a science of social work. Research on Social Work Practice, 1049731512441263.

 

Rice, E., Petering, R., Stringfellow, E., & Craddock, J. (2017). Innovations in Community-Based and Interdisciplinary Research. Research on Social Work Practice.

 

Tambe: Introduction to Artificial Intelligence

Readings:

100 year study on Artificial Intelligence (AI100):

https://ai100.stanford.edu/2016-report

Download: https://ai100.stanford.edu/sites/default/files/ai100report10032016fnl_singles.pdf

 

“AI for Social Good” workshop videos and workshop report:

http://cra.org/ccc/artificial-intelligence-social-good-speakers/

 

Rice/Tambe: An example Social Work + AI project

In both cases, we will introduce what is the history of the field, what are some major schools of thought, what are current trends. We will introduce basics, things taken for granted. For example, experiment design, IRB approval, search…

A key topic of this class will be to get students to form teams. Teams must include students from social work and AI; while teaming arrangements could change over time, the initial team will be responsible for delivering potential topics for book chapters by Feb 6.

 

Jan 16:

Holiday (MLK day)

 

 

 

 

 

Jan 23:

Rice: Introduction to popular opinion leader interventions

Required Readings:

Kelly, J. A., St Lawrence, J. S., Diaz, Y. E., Stevenson, L. Y., Hauth, A. C., Brasfield, T. L., ... & Andrew, M. E. (1991). HIV risk behavior reduction following intervention with key opinion leaders of population: an experimental analysis. American Journal of Public Health, 81(2), 168-171.

 

Valente, T. W. (2012). Network interventions. Science, 337(6090), 49-53.

 

Schneider, J. A., Zhou, A. N., & Laumann, E. O. (2015). A new HIV prevention network approach: sociometric peer change agent selection. Social Science & Medicine, 125, 192-202.

 

Recommended Readings:

Whitbeck, L. B., & Hoyt, D. R. (1999). Nowhere to grow: Homeless and runaway adolescents and their families. Transaction Publishers., ch 1.

 

Milburn, N. G., Rice, E., RotheramBorus, M. J., Mallett, S., Rosenthal, D., Batterham, P., ... & Duan, N. (2009). Adolescents exiting homelessness over two years: The risk amplification and abatement model. Journal of research on adolescence, 19(4), 762-785.

 

 

Tambe: Introduction to decision theory, risk averseness, MDP, POMDP (sequential planning under uncertainty), Social influence maximization, POMDP application to social network intervention

 

            Required Readings:

Decision Theory handout emailed to students.

 

Amulya Yadav , Leandro Marcolino, Eric Rice, Robin Petering, Hailey Winetrobe, Harmony Rhoades, Milind Tambe, Heather Carmichael Preventing HIV Spread in Homeless Populations using PSINET In Proceedings of the Annual Conference on Innovative Applications of Artificial Intelligence (IAAI) 2015

http://teamcore.usc.edu/people/SocialGood/index.html

 

Recommended Readings:

Amulya yadav, Hau Chan, Albert Jiang, Haifeng Xu, Eric Rice, Milind Tambe Using Social Networks to Aid Homeless Shelters: Dynamic Influence Maximization Under Uncertainty In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS) 2016

 

Jan 30

Rice: Introduction to diffusion of innovations theory to motivate social network interventions

Required Readings:

Rogers, E. M. (2010). Diffusion of innovations. Simon and Schuster., Ch1, pp 1-33. “Elements of Diffusion”

 

Christakis, N. A., & Fowler, J. H. (2013). Social contagion theory: examining dynamic social networks and human behavior. Statistics in medicine, 32(4), 556-577.

 

Recommended Readings:

Rogers, E. M. (2002). Diffusion of preventive innovations. Addictive behaviors, 27(6), 989-993.

 

McPherson, M., Smith-Lovin, L., & Cook, J. M. (2001). Birds of a feather: Homophily in social networks. Annual review of sociology, 415-444.

 

Christakis, N. A., & Fowler, J. H. (2007). The spread of obesity in a large social network over 32 years. New England journal of medicine, 357(4), 370-379.

 

Tambe: Introduction to security games and network security games; basic game theory, Normal form games, Dominance, Nash equilibrium; Mixed strategy Nash equilibrium, Stackelberg games

            Required Readings:

            Handout: Game Theory

Serious games, T Nguyen, Significance 13 (6), 14-19, 2016

Ariel D. Procaccia “Game Theory Is Useful, Except When It Is Not” http://www.symposium-magazine.com/game-theory-is-useful-except-when-it-is-not-ariel-d-procaccia/

Thanh H. Nguyen, Debarun Kar, Matthew Brown, Arunesh Sinha, Albert Xin Jiang, Milind Tambe “Towards a Science of Security Games” New Frontiers of Multidisciplinary Research in STEAM-H (Book chapter) (edited by B Toni), 2016

http://teamcore.usc.edu/pubDetails.aspx?id=792

 

Recommended Readings:

Benjamin Ford, Matthew Brown, Amulya Yadav, Amandeep Singh, Arunesh Sinha, Biplav Srivastava, Christopher Kiekintveld, and Milind Tambe  “Protecting the NECTAR of the Ganga River through Game-Theoretic Factory Inspections” International Conference on Practical Applications of Agents and Multi-Agent Systems (PAAMS), 2016

 

Jan 30 (Monday) Book chapter/paper proposal due

 

Feb 6

Guest Lecture (AI): Prof. Phebe Vayanos

Guest Lecture (Social Work): Prof. Shinyi Wu

Book Chapter/Paper proposals due

 

 

Feb 13

Rice: Introduction to substance abuse prevention

Required Readings:

Rice, E., Milburn, N. G., & Monro, W. (2011). Social networking technology, social network composition, and reductions in substance use among homeless adolescents. Prevention Science, 12(1), 80-88.

Najavits, L. M., Weiss, R. D., Shaw, S. R., & Muenz, L. R. (1998). “Seeking safety”: Outcome of a new cognitivebehavioral psychotherapy for women with posttraumatic stress disorder and substance dependence. Journal of traumatic stress, 11(3), 437-456.

Recommended Readings:

Barman-Adhikari, A., Rice, E., Winetrobe, H., & Petering, R. (2015). Social network correlates of methamphetamine, heroin, and cocaine use in a sociometric network of homeless youth. Journal of the Society for Social Work and Research, 6(3), 433-457.

Desai, R. A., Harpaz-Rotem, I., Najavits, L. M., & Rosenheck, R. A. (2008). Impact of the seeking safety program on clinical outcomes among homeless female veterans with psychiatric disorders. Psychiatric services, 59(9), 996.

 

Tambe: Security and game theory, network security games

 

Required Readings:

Jason Tsai, Thanh H. Nguyen, Milind Tambe Security Games for Controlling ContagionConference on Artificial Intelligence (AAAI), 2012

J. Pita, M. Jain, C. Western, P. Paruchuri, J. Marecki, M. Tambe, F. Ordonez, S. Kraus Using game theory for Los Angeles Airport Security In AI Magazine 30(1):43-57, 2009.

B. An, E. Shieh, R. Yang, M. Tambe, C. Baldwin, J. DiRenzo, B. Maule, G. Meyer PROTECT – A Deployed Game Theoretic System for Strategic Security Allocation for the United States Coast Guard In AI Magazine, 33(4):96-110, 2012.

 

Recommended Readings:

C. Kiekintveld, J. Pita, M. Jain, J. Tsai, M. Tambe, F. Ordonez Computing Optimal Randomized Resource Allocations for Massive Security Games In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2009

P. Paruchuri, J. Pearce, J. Marecki, M. Tambe, F. Ordonez, S. Kraus Playing games with security: An Efficient Exact Algorithm for Bayesian Stackelberg Games In Proceedings of the Seventh International Joint Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2008

F. Fang, T. Nguyen1, R. Pickles, W.Y. Lam, G.R. Clements2, B. An, A. Singh, B.C. Schwedock, M. Tambe, A. Lemieux PAWS A Deployed Game-Theoretic Application to Combat Poaching In AI Magazine (to appear)

 

 

February 17 (Friday) — Problem statement/introduction portion of chapters due

 

Feb 20: Holiday (President’s day)

 

Feb 27

Tambe: Introduction to Machine learning, neural nets, decision trees and crime

Required:

Handout on machine learning

Recommended Reading:

Eric Potash, Joe Brew, Alexander Loewi, Subhabrata Majumdar, Andrew Reece, Joe Walsh, Eric Rozier, Emile Jorgenson, Raed Mansour, Rayid Ghani. Predictive Modeling for Public Health: Preventing Childhood Lead Poisoning. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

Handout:http://www.nytimes.com/2016/06/08/technology/online-searches-can-identify-cancer-victims-study-finds.html

Reading: M.J. Paul, R.W. White, E. Horvitz. Diagnoses, Decisions and Outcomes: Web Search as a Decision Support for Cancer. In Proceedings of WWW 2015

Recommended:

Stefano Ermon et al, paper: http://science.sciencemag.org/content/353/6301/790 video: https://www.youtube.com/watch?v=DafZSeIGLNE&feature=youtu.be

 

Rice: experimental design

Required readings:

To be announced

 

 

 

March 6

Rice: Social science approaches to agent-based simulations of disease and other social problems

Required Readings:

Axelrod, R. M. (2006). The evolution of cooperation. Basic books. (ch 1)

Additional to be announced

Recommended Readings:

Nowak, M. A. (2006). Five rules for the evolution of cooperation. science, 314(5805), 1560-1563.

 

Tambe: Agent-based simulations and their applications

M. Tambe, W. L. Johnson, R. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Scwhamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1):15–39, 1995.

J.Tsai, N.Fridman, E.Bowring,M.Brown,S.Epstein,G.Kaminka,S.Marsella,A.Ogden,I.Rika, A. Sheel, M. Taylor, X. Wang, A. Zilka, M. Tambe ESCAPES: Evacuation Simulation with Children, Authorities,Parents,Emotions,andSocial Comparison In Proceedings of the International Conference on Autonomous Agents and Multiagent Systems (AAMAS) (Innovative Applications Track), May 2011

J. Tsai, E. Bowring, S. Marsella, M.Tambe Empirical Evaluation of Computational Emotional Contagion Models In Proceedings of the International Conference on Intelligent Virtual Agents(IVA), September 2011

 

 

 

March 10 (Friday) — Preliminary results write up for chapters due

 

 

March 13: Holiday (Spring Recess)

 

March 20

Rice: Social science approaches to decision-making under uncertainty and social problems driven by uncertainty

Required readings:

Cook, K. S., Cheshire, C., Rice, E. R., & Nakagawa, S. (2013). Social exchange theory. In Handbook of social psychology (pp. 61-88). Springer Netherlands.

Fulginiti, A., Pahwa, R., Frey, L. M., Rice, E., & Brekke, J. S. (2015). What factors influence the decision to share suicidal thoughts? A multilevel social network analysis of disclosure among individuals with serious mental illness. Suicide and Life-Threatening Behavior.

 

Tambe: Human behavior modeling and emotions

Required readings:

Handout on behavioral game theory

R. Yang, C. Kiekintveld, R. John, F. Ordonez, M. Tambe Improving Resource Allocation Strategy Against Human Adversaries in Security Games In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), July 2011

 

Recommended readings:

J. Pita, R. Maheswaran, M. Tambe, S. Kraus A Robust Approach to Addressing Human Adversaries in Security Games In Proceedings of the European Conference on Artificial Intelligence (ECAI), August 2012.

T. Nguyen, R. Yang, A. Azaria, S. Kraus, M. Tambe Analyzing the effectiveness of adversary modeling in security games In Proceedings of the Conference on Artificial Intelligence (AAAI), August 2013.

D. Kar, F. Fang, F.M. Delle Fave, N. Sintov, M. Tambe, ”A Game of Thrones”: When Human Behavior Models Compete in Repeated Stackelberg Security Games In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May, 2015

 

 

 

March 27

Guest Lecture (Social Work): Prof. Daniel Hackman

 

 

April 3

The ethics of AI for social good and distopian views on AI

Guest Lecture: To be announced

Readings: TBD

April 3 (Monday) Full first draft of chapters due

 

April 10

Guest Lecture (AI): To be announced

Guest Lecture (Social Work): To be announced

 

April 17

Guest Lecture (AI): To be announced

Guest Lecture (Social Work): To be announced

 

April 24

Final project presentations


 

 

Class Expectations:

 

Group Assignment: Book Chapter (90% of class grade)

A complete draft, ready for external review of a chapter focussed on applying artificial intelligence techniques to a critical social problem  due at the end of the semester.  These chapters will be included in a book which Drs. Tambe, Rice and Vayanos are editing.  Groups must work collaboratively and include students from both engineering and social work.  A statement of author contributions (i.e. who did what) must be turned in with the final draft. 

 

Rough drafts and partial drafts will be due at different points throughout the semester so that we may provide students with constructive criticism along the way. 

5% proposal - Due Monday Jan 30

15% motivate the problem - Due Friday February 17

10% prelim results - Due Friday March 10

15% initial full draft - Due Monday April 3

45% final paper - Due Friday May 5

Ideally chapters will include results from simulation models or modeling techniques using existing real world data.  These chapters may be “aspirational”.  An actual intervention is not possible in the time frame of one semester, but groups who propose intervention strategies should have preliminary hypothetical models, at a minimum.

 

Class Participation (10% of Course Grade)

Students’ active involvement in the class is considered essential to their growth as practitioners. Their presence in class, along with preparation by having read and considered the assignments, and participation in discussion and activities are essential.


 

 

 

Statement on Academic Conduct and Support Systems

 

Academic Conduct

Plagiarism – presenting someone else’s ideas as your own, either verbatim or recast in your own words – is a serious academic offense with serious consequences.  Please familiarize yourself with the discussion of plagiarism in SCampus in Section 11, Behavior Violating University Standardshttps://scampus.usc.edu/1100-behavior-violating-university-standards-and-appropriate-sanctions/.  Other forms of academic dishonesty are equally unacceptable.  See additional information in SCampus and university policies on scientific misconduct, http://policy.usc.edu/scientific-misconduct/.

 

Discrimination, sexual assault, and harassment are not tolerated by the university.  You are encouraged to report any incidents to the Office of Equity and Diversity http://equity.usc.edu/ or to the Department of Public Safety http://capsnet.usc.edu/department/department-public-safety/online-forms/contact-us.  This is important for the safety whole USC community.  Another member of the university community – such as a friend, classmate, advisor, or faculty member – can help initiate the report, or can initiate the report on behalf of another person.  The Center for Women and Men http://www.usc.edu/student-affairs/cwm/ provides 24/7 confidential support, and the sexual assault resource center webpage sarc@usc.edu describes reporting options and other resources.

 

Support Systems

A number of USC’s schools provide support for students who need help with scholarly writing.  Check with your advisor or program staff to find out more.  Students whose primary language is not English should check with the American Language Institute http://dornsife.usc.edu/ali, which sponsors courses and workshops specifically for international graduate students.  The Office of Disability Services and Programs http://sait.usc.edu/academicsupport/centerprograms/dsp/home_index.htmlprovides certification for students with disabilities and helps arrange the relevant accommodations.  If an officially  declared emergency makes travel to campus infeasible, USC Emergency Information http://emergency.usc.edu/will provide safety and other updates, including ways in which instruction will be continued by means of blackboard, teleconferencing, and other technology.