AAAI-19 will feature an emerging track on artificial intelligence for social impact. The track recognizes that high-quality research on social impact domains often leads to papers that differ from traditional AAAI submissions along multiple dimensions. These are motivated by the following issues:

  • Data collection may be difficult and may require innovative methods and validations, for instance to address large scale data gathering in the field, eliminate bias and ensure fairness;
  • Problem modeling is a time-intensive activity that require significant collaborations with domain experts and need to balance a variety of tradeoffs in decision making;
  • Social impact is only realized through time-consuming field studies that typically compare a baseline with the application of novel algorithms in the real world.

The goal of this track at AAAI 2019 is to highlight these technical challenges and opportunities and to showcase the social benefits of artificial intelligence.

Submissions will follow the regular AAAI technical paper submission process. The AAAI-19 call for papers features a category “AI for Social Impact” for identifying the papers submitted to this track. We encourage the use of other keywords to describe paper content more comprehensively.

Papers submitted to this track will be evaluated using the following criteria. We note that it is virtually impossible for any paper to score highly in all dimensions, and so encourage authors who have concentrated on a subset of these dimensions to submit their work.

Novelty of problem

  1. The social impact problem considered by this paper was entirely novel
  2. The social impact problem considered by this paper was previously studied but not in the AI literature
  3. This paper represents a new take on a problem that has been considered in the AI community before
  4. This paper’s contribution was elsewhere: it follows up on an existing problem formulation

Engagement with literature

  1. Shows an excellent understanding of other literature on the problem, including that outside computer science
  2. Shows a strong understanding of other literature on the problem, perhaps focusing on various substrands or on the CS literature
  3. Shows a moderate understanding of other literature on the topic, but does not engage in depth
  4. Does not engage sufficiently with other literature on the topic

Novelty of approach

  1. Introduces a new model, data gathering technique, algorithm, and/or data analysis technique
  2. Substantially improves upon an existing model, data gathering technique, algorithm, and/or data analysis technique
  3. Makes a moderate improvement to an existing model, data gathering technique, algorithm, and/or data analysis technique
  4. This paper’s contribution was elsewhere: it employs existing models, data gathering techniques, algorithms, and data analysis techniques

Justification of approach

  1. Thoroughly and convincingly justifies the approach taken, explaining strengths and weaknesses as compared to other alternatives
  2. The justification of the approach is convincing overall, but could have been more thorough and/or alternatives could have been considered in more detail
  3. The justification of the approach is relatively convincing, but has weaknesses
  4. The justification of the approach is flawed and/or not convincing

Quality of evaluation

  1. Evaluation was exemplary: data described the real world and was analyzed thoroughly
  2. Evaluation was convincing: datasets were realistic; analysis was solid
  3. Evaluation was adequate, but had significant flaws: datasets were unrealistic and/or analysis was insufficient
  4. Evaluation was unconvincing

Facilitation of follow-up work

  1. Excellent facilitation of follow-up work: open-source code; public datasets; and a very clear description of how to use these elements in practice
  2. Strong facilitation of follow-up work: some elements are shared publicly (data, code, or a running system) and little effort would be required to replicate the results or apply them to a new domain
  3. Adequate facilitation of follow-up work: moderate effort would be required to replicate the results or apply them to a new domain
  4. Weak facilitation of follow-up work: considerable effort would be required to replicate the results or apply them to a new domain

Scope and promise for social impact

  1. Likelihood of social impact is extremely high: the paper’s ideas are already being used in practice or could be immediately
  2. Likelihood of social impact is high: relatively little effort would be required to put this paper’s ideas into practice, at least for a pilot study
  3. Likelihood of social impact is moderate: this paper gets us closer to its goal, but considerably more work would be required before the paper’s ideas could be implemented in practice
  4. Likelihood of social impact is low: the ideas proposed in this paper are unlikely to make a significant impact on the proposed problem

Reliance upon and/or advancement of cutting edge AI techniques

  1. Introduced novel AI techniques suited to the problem being solved
  2. Relied upon state-of-the-art AI techniques
  3. Relied upon relatively standard AI techniques
  4. It is not clear why this paper would appear at an AI conference.

Overall recommendation:

  1. Certainly accept
  2. Probably accept
  3. Borderline accept
  4. Borderline reject
  5. Probably reject
  6. Certainly reject

Accepted papers will be presented at the main conference in sessions designated to the topic.

Call for Papers

Emerging Topics Cochairs

  • Kevin Leyton-Brown (University of British Columbia, Canada)
  • Milind Tambe (University of Southern California, USA)

This site is protected by copyright and trademark laws under US and International law. All rights reserved. Copyright © 1995–2018 AAAI