Announcing the winners of the Facebook Mechanism Design for Social Good research awards

We are pleased to announce the winners of the Facebook Mechanism Design for Social Good research awards. In this request for proposals, we asked researchers to consider the following problem: suppose there is (1) an existing online platform that is actively used by the population, and (2) an existing set of social ills (e.g., unemployment, disease, poverty, divisiveness, loneliness). How should one design mechanisms on top of such an online platform to build community in a way that alleviates these social ills?

Below are the winning proposals:

  • Incentives on Scale, Mechanisms for crowdsourcing with small-holder farmers
    PI: Mutembesa Daniel, Makerere University
    Collaborator: Boi Faltings, EPFL
    Collaborator: Christopher Omongo, National Crops Resources and Research Institute
    Collaborator: Humphrey Mutaasa, Uganda National Farmers Federation
  • Modern Social Choice: Mechanisms and Platforms for Large Scale Deliberation
    PI: Ashish Goel, Stanford University
    Co-PI: James S. Fishkin, Stanford University
    Co-PI: Kamesh Munagala, Duke University
  • Promoting Diversity in Peer Production through Mechanism Design
    PI: Zhiwei Steven Wu, University of Minnesota
    Co-PI: Haiyi Zhu, University of Minnesota

Details about each of these projects are below.

Incentives on Scale, Mechanisms for crowdsourcing with small-holder farmers
Consider a common problem faced by farmers in the developing world. The farmer relies on a healthy crop to provide for his or her family, but that crop is continually at risk of being destroyed: diseases and pests can ruin the farmer’s entire harvest, and outbreaks can affect the broader farming community. Such dangers are hard to detect in their early stages, and so farmers rely on agricultural experts, who travel across the country, identifying damaged crops and recommending ways for farmers to mitigate their losses. While farmers benefit from these insights, experts are limited in number, and their coverage is restricted to well-traveled roads. A farmer may lose precious time while waiting for an expert to arrive, or not receive any such guidance at all, which could mean the difference in whether or not the farmer’s crop survives.

In their previous work, The PIs gave farmers phones with high resolution cameras, incentivized the farmers to photograph their crops, and then used those images to identify diseases and outbreaks via machine learning models. The effectiveness of such an approach requires that pictures are of sufficiently high quality and are taken in relevant locations. The PIs have thus far taken an ad-hoc approach in deciding how to incentivize farmers: they tried various incentive schemes, but not systematically.

In this proposal, the PIs will scale up their past initiative (i.e., get more farmers collecting data) and run randomized control trials to understand how different incentive schemes affect the quality of data received for identifying trends in crop disease.

Modern Social Choice: Mechanisms and Platforms for Large Scale Deliberation
The goal of this work is to help citizens make informed decisions about public policy issues. Currently, political discussions are often divisive and polarized, and resort to name-calling rather than discussing and considering reasoned arguments from the other side.

The authors have spent a long time working on deliberative polling. The main idea of deliberative polling is to choose a representative sample of the population and get them to have moderated discussions in small groups. The results of the process are then analyzed and presented to the public.

In this proposal, the authors propose to scale up their past work in deliberative polling. To do so, they need to automate some of the sampling and moderation processes. They plan to create an online bot that monitors conversations and steers them in the appropriate direction, and consider ways for participants to express when conversation has derailed. They plan to run experiments to better understand which design is most effective (e.g., “Does a visible speaker queuing system for speaking increase or decrease engagement and civility? Does automated flagging of offensive comments using NLP result in more or less civility in the conversation as compared to a system where group members collectively flag offensive comments using a simple interface?”).

Additionally, they plan to better understand how to get a representative population through ad targeting (soft targeting), how to allow participants to express preferences in complex settings (i.e., not just a binary vote) and how to aggregate those preferences.

Promoting Diversity in Peer Production through Mechanism Design
In many online platforms, the producers and consumers of content are not representative of each other demographically. For example, in Wikipedia and open street maps, significantly more men contribute to content creation than women. This results in a lack of diversity in content; for example, a relative lack of articles about women mathematicians.

The authors propose to study how a social referrals mechanism can lead to more diverse content creation. The idea is to incentivize people to refer an expert to write or revise an article. From an initial set of seed candidates, referrals will propagate outwards until an expert is found. Budgets are adjusted dynamically. Content quality will be incentivized through badges, considering quality, quantity of submissions, and dangers of burnout.

Conclusion and next steps

The selection committee was extremely impressed by the large number of top-quality submissions received. While we could only provide $50,000 in funding for three submissions, many more of the 58 submissions we received contained exciting research directions. We reviewed inspiring research proposals in domains such as refugee resettlement, blood donations, healthcare, civic engagement, charitable donations, misinformation, innovation contests, disaster evacuation and relief, agricultural markets, dating markets, general matching markets and many more. We appreciate the time and effort so many spent writing proposals, and wish all submission authors the best of luck in pursuing their research agendas.

Additionally, we especially want to call out the Mechanism Design for Social Good organization, which is an independent group but was the inspiration for this award. The MD4SG organization has an upcoming workshop, as well as a regular colloquium series and working groups. See their website for more information.

Facebook has a number of teams working in social good. If you have expertise in machine learning, AI or economics and computation, and are interested in applying that expertise to social good causes, please apply to work with us. We’re hiring interns, postdocs, full-time employees and academic contractors.

The post Announcing the winners of the Facebook Mechanism Design for Social Good research awards appeared first on Facebook Research.


We need your support to operate it properly. We have lots of ideas but less fund, so help us with your funding.


Thoughts of the day

A perfection of means, and confusion of aims, seems to be our main problem

Albert Einstein
Sell your product online

Do you want to sell products online with no extra cost?

Send your details, our executive will contact you




What you like most in facebook?








Tutorial On Request
Q. Ask us for any tutorial or any thing which helps to build your career better.