Schedule

All of the papers should be available through google scholar. Logging into Harvard library should also make it easy to access these papers. 

Definitions, course structure:

1. (Jan 26) Lecture 1: Course intro, syllabus, goals, key concepts: Perspectives on AI in Society; overview. 

Lecture 1 slides

 

Handout: Syllabus

Optional readings:

 

AI for social good case studies, philosophy:

2. (Jan 28) Lecture 2: AI for social impact case studies and discussion

 

Lecture video discussing AI and public safety 

Lecture 2 slides

 

3. (Feb 2) Lecture 3: AI for social impact case studies and discussion

 

Lecture video discussing AI for public health and conservation

Lecture 3 Slides

 

The goal of these two lectures is to explain my own group’s previous projects over the past 15 years in AI for Social Impact. I will discuss key projects, how they came about, what was the motivation, their progress, obstacles, lessons learned at the end. 

 

Optional readings:

Background readings:

  • D. Kempe, J. Kleinberg, and É. Tardos. Maximizing the Spread of Influence through a Social Network. In Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining, pages 137–146. ACM, 2003. 

  • H. Kamarthi, P. Vijayan, B. Wilder, B. Ravindran,M. Tambe Influence maximization in unknown social networks: Learning Policies for Effective Graph Sampling In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May, 2020

 

4. (Feb 4) Lecture 4: AI for Substance Abuse prevention (Invited lecture by Prof. Amulya Yadav)

 

 

Required readings

  • Maryam Tabar, Heesoo Park, Stephanie Winkler, Dongwon Lee, Anamika Barman-Adhikari, Amulya Yadav “Identifying Homeless Youth At-Risk of Substance Use Disorder: Data-Driven Insights for Policymakers” KDD 2020
  • Amulya Yadav, Roopali Singh, Nikolas Siapoutis, Anamika Barman-Adhikari, Yu Liang “Optimal and Non-Discriminative Rehabilitation Program Design for Opioid Addiction Among Homeless Youth” IJCAI 2020
  • Zi-Yi Dou, Anamika Barman-Adhikari, Fei Fang, Amulya Yadav “Harnessing Social Media to Identify Homeless Youth At-Risk of Substance Use”, AAAI 2021

Optional 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.
  • Thomas W Valente, Beth R Hoffman, Annamara Ritt-Olson, Kara Lichtman, and C Anderson Johnson. 2003. Effects of a social-network method for group assignment strategies on peer-led tobacco prevention programs in schools. American journal of public health 93, 11 (2003), 1837–1843.

 

Ethics of AI for Social Good

5. (Feb 9) Lecture 5:  AI & Ethics (Embedded ethics)

 

LECTURE 5 SLIDES

 

Ethical reasoning is an essential skill for today’s computer scientists. The Embedded EthiCS distributed pedagogy embeds philosophers directly into computer science courses to teach students how to think through the ethical and social implications of their work.

  •  Thomas W Valente, Beth R Hoffman, Annamara Ritt-Olson, Kara Lichtman, and C Anderson Johnson. 2003. Effects of a social-network method for group assignment strategies on peer-led tobacco prevention programs in schools. American journal of public health 93, 11 (2003), 1837–1843.
  • A. Rahmattalabi, A. Yadav, B. Wilder, A. Fulginiti, P. Vayanos, E. Rice, M. Tambe Exploring Algorithmic Fairness in Robust Graph Covering Problems In Proceedings Conference on Neural Information Processing Systems (NeurIPS), December, 2019

 

Defining projects

6. (Feb 11) Lecture 6: Prof. Chris Golden: Public health and conservation

  • Sarah Whitmee, Andy Haines, Chris Beyrer, Frederick Boltz, Anthony G Capon, Braulio Ferreira de Souza Dias, Alex Ezeh, Howard Frumkin, Peng Gong, Peter Head, Richard Horton, Georgina M Mace, Robert Marten, Samuel S Myers, Sania Nishtar, Steven A Osofsky, Subhrendu K Pattanayak, Montira J Pongsiri, Cristina Romanelli, Agnes Soucat, Jeanette Vega, Derek Yach “Safeguarding human health in the Anthropocene epoch:report of The Rockefeller Foundation–Lancet Commission on planetary health” The Rockefeller Foundation–Lancet Commission on planetary health

 

7. (Feb 16) Lecture 7: WWF Invited lecture (Rohit Singh)

WWF LECTURE SLIDES

Background on Tuesday's WWF lecturer, Rohit Singh. 

https://wwf.panda.org/discover/knowledge_hub/?356370/Life-on-the-Frontline-2019-A-global-survey-of-the-working-conditions-of-rangers (Links to an external site.) 

https://tigers.panda.org/reports/ (Links to an external site.)

https://wwf.panda.org/discover/our_focus/wildlife_practice/wildlife_trade/wildlife_crime_initiative/

 

8. (Feb 18) Lecture 8: AI and COVID-19: debate. Role of in the COVID-19 fight?

Pointer to Michael Mina lecture at CRCS

 

Prof. Michael Mina (invited lecture)

  • B. Wilder, M. Charpignon, J. Killian, H. Ou, A. Mate, S. Jabbari, A. Perrault, A. Desai, M. Tambe,  M. Majumder “Modeling between-population variation in COVID-19 dynamics in Hubei, Lombardy, and New York City” In Proceedings of the National Academy of Sciences (PNAS), 117(41): 25904-25910, 2020.
  • D. Larremore, B. Wilder, E. Lester, S. Shehata, J. Burke, J. Hay, M. Tambe, M. Mina, R. Parke “Test sensitivity is secondary to frequency and turnaround time for COVID-19 surveillance” In Science Advances, eabd5393, 2020

 

 

9. (Feb 23) Lecture 9: Discussion AI for Social Good; plus bandit models for health intervention 

 

Discussion (available on most streaming services):

POVERTY INC

 

Papers:

  • A. Mate*, J. Killian*, H. Xu, A. Perrault, M. Tambe (* equal contribution) Collapsing Bandits and Their Application to Public Health Interventions In Proceedings Conference on Neural Information Processing Systems (NeurIPS), December, 2020 
  • A. Biswas, G. Aggarwal, P. Varakantham, M. Tambe Learning Index Policies for Restless Bandits with Application to Maternal Healthcare In International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2021

Optional:

  • P. Whittle. Restless bandits: Activity allocation in a changing world. J. Appl. Probab., 25(A): 287–298, 1988.
  • R. R. Weber and G. Weiss. On an index policy for restless bandits. J. Appl. Probab., 27(3): 637–648, 1990. 

 

10. (Feb 25) Lecture 10: Implementation Science by Prof. Shoba Ramanadhan 

 

Shoba Ramanadhan Lecture Slides on implementation science

 

https://pubmed.ncbi.nlm.nih.gov/31036287/

  • Mark S. Bauer, Laura Damschroder, Hildi Hagedorn, Jeffrey Smith & Amy M. Kilbourne “An introduction to implementation science for the non-specialist” BMC Psychology volume 3, Article number: 32 (2015)

 

 

STUDENT PAPER PRESENTATIONS

 

11. (Mar 2) Lecture 11: Student paper presentation

Paper: Project RISE: Recognizing Industrial Smoke Emissions

Appears at: AAAI 2021 (to appear), Special Track on AI for Social Impact

 

 RainBench: Towards Global Precipitation Forecasting from Satellite Imagery

Appears at:

 

Paper: A Markov Decision Process Model for Socio-Economic Systems Impacted by Climate Change

Appears at: ICML 2020

  

Paper: Security and privacy in Smart Farming: Challenges and Opportunities

Appears at: IEEE Access 8 (2020)

 

12. (Mar 4) Lecture 12: Student presentations 

Paper: Dual-Mandate Patrols: Multi-Armed Bandits for Green Security

Appears at: AAAI-21

 

Paper: Protecting Geolocation Privacy of Photo Collections

Appears at: AAAI-2020

 

Paper: Subverting Privacy-Preserving GANs: Hiding Secrets in Sanitized Images

Appears at: AAAI-21 - Special Track on AI for Social Impact

 

Paper: Weakly-Supervised Fine-Grained Event Recognition on Social Media Texts for Disaster Management

Appears at: AAAI 2020 Special Technical Track: AI for Social Impact

 

13. (Mar 9) Lecture 13: Student presentations

Paper: Predicting Patient Outcomes with Graph Representation Learning

Appears at: AAAI 2021

  

Paper: Personalizing ASR for Dysarthric and Accented Speech with Limited Data

Appears at: Interspeech 2019

  

Paper: Using Radio Archives for Low-Resource Speech Recognition: Towards an Intelligent Virtual Assistant for Illiterate Users

Appears at: AAAI 2021 Special Track on AI for Social Impact

 

14. (Mar 11) Lecture 14:  Midterm short presentations on project progress

 

 

March 16: Wellness day no courses meet

AI and sustainability: Learning to appreciate an interdisciplinary perspective

15. (Mar 18) Lecture 15:Invited Lecture on AI and Sustainability by Prof. Andrew Davies, Harvard EOB

Andrew Davies Lecture Slides 

Brodrick PG, Davies AB, and Asner GP. 8/2019. “Uncovering ecological patterns with convolutional neural networks.Links to an external site.” Trends in Ecology and Evolution, 34, 8, Pp. 734-745. Publisher's Version

 

 

From Data to Deployment in AI4SG

16. (Mar 23) Lecture 16:  Invited lecture by Bryan Wilder on measuring impact

BRYAN WILDER SLIDES ON MEASURING IMPACT

 

 

17. (Mar 25) Lecture 17: Discuss broader impacts of class projects

 

M. Latonero “Opinion: AI For Good Is Often Bad” Wired November 2019

The conversation “AI algorithms intended to root out welfare fraud often end up punishing the poor instead”, Rawstory Feb 14, 2020

Beyond ‘AI for Social Good’ (AI4SG): social transformations—not tech-fixes—for health equity
Cheryl Holzmeyer, Institute for Social Transformation, University of California, Santa Cruz, CA, USA

E. Gibney “The battle for ethical AI at the world’s biggest machine-learning conference” Nature news, January 2020

 

18. (Mar 30) Lecture 18: Invited Lecture by Phil Nelson, Google

 

 

Fairness, Accountability, Transparency

19. (Apr 1) Lectures 19: Fairness in AI for Social Good

 

FAIRNESS & AI FOR SOCIAL GOOD

 

A. Rahmattalabi, A. Yadav, B. Wilder, A. Fulginiti, P. Vayanos, E. Rice, M. Tambe Exploring Algorithmic Fairness in Robust Graph Covering Problems In Proceedings Conference on Neural Information Processing Systems (NeurIPS), December, 2019

A. Rahmattalabi, S. Jabbari, P. Vayanos, H. Lakkaraju, M. Tambe Fair Influence Maximization: a Welfare Optimization Approach In In AAAI conference on Artificial Intelligence (AAAI), February, 2021

 

 

20. (Apr 6) Lecture 20: Question-answer session on second round of paper readings

 

Paper: GLTR: Statistical Detection and Visualization of Generated Text

Appears at: ACL 2019 Demo Track

 

Paper: A Distributed Multi-Sensor Machine Learning Approach to Earthquake Early Warning

Appeared at: AAAI 2020

 

Paper: Predicting Forest Fire Using Remote Sensing Data And Machine Learning

Appeared at: AAAI 2021, Special Track on AI for Social Impact

 

Paper: Combining satellite imagery and machine learning to predict poverty

Appears at: Science, Vol. 353, Issue 6301, 2016

 

 

Paper: Inferring Nighttime Satellite Imagery from Human Mobility

Appears at: AAAI 2020 Special Track on AI for Social Impact

 

 

Paper: Feature exploration for almost zero-resource ASR-free keyword spotting using a multilingual bottleneck extractor and correspondence autoencoders

Appears at: Interspeech 2019

 

Paper: Minimizing Energy Use of Mixed-Fleet Public Transit for Fixed-Route Service

Appears at: AAAI 2021

 

 

Paper: Spatio-Temporal Attention-Based Neural Network for Credit Card Fraud Detection

Appears at: AAAI-2021, Special Track on AI for Social Impact

 

 

Paper: Characterizing soundscapes across diverse ecosystems using a universal acoustic feature set

Appears at: PNAS 2020

 

 

Paper: How Robust are the Estimated Effects of Nonpharmaceutical Interventions against COVID-19?

Appears at: NeurIPS 2020

 

21. (Apr 8) Lecture 21: AI for Social Good and HCI

 

  1. Enabling Data-Driven API Design with Community Usage Data: A Need-Finding Study
    Tianyi Zhang, Björn Hartmann, Miryung Kim, and Elena Glassman. CHI 2020.

  2. Community power could boost confidence in vaccination programmes
    RC Wurth, H Saksono
    Nature 589 (7841), 198

  3. Storywell: Designing for Family Fitness App Motivation by Using Social Rewards and Reflection
    Herman Saksono, Carmen Castaneda-Sceppa, Jessica Hoffman, Vivien Morris, Magy Seif El-Nasr, Andrea Parker. 2020. Storywell: Designing for Family Fitness App Motivation by Using Social Rewards and Reflection. In CHI Conference on Human Factors in Computing Systems Proceedings (CHI 2020), May 4–9, 2019, Honolulu, HI, USA. ACM, New York, NY, USA, 13 pages. (PDF)3.

  4. Social technologies for digital wellbeing among marginalized communities
    MA Devito, AM Walker, J Birnholtz, K Ringland, K Macapagal, A Kraus, ...
    Conference Companion Publication of the 2019 on Computer Supported

 

22. (Apr 13) Lecture 22: Interpretability in machine learning Invited Lecture by Dr.  Hima Lakkaraju 

Hima Lakkaraju Lecture Slides on Explainability in ML

 

Apr 15: wellness day

23. (Apr 20) Lecture 23: Final project presentations
 
24. (Apr 22) Lecture 24: Final project presentations

25. (Apr 27) Lecture 25: Final project presentation