Evelyne Viegas

Involving students and trainees both in organizing and in participating in competitions and challenges is a powerful pedagogical tool. From kindergarten through the last years of graduate studies, or as a way to upskill to gain new skills on the job, competitions can gamify the learning process and thus motivate people to explore by themselves, assimilate a variety of material, and expand their capabilities. Competitions also contribute to engaging students in learning about problems of societal importance. At the graduate level, designing and implementing competitions can be seen as a hands-on means of learning proper design and analysis of experiments. In the workplace, this remains an effective way to embrace continuous learning and remain up to date as technologies, tools, and approaches evolve. In this chapter, we present a synthesis of various hands-on teaching and learning experiences using competitions as a medium. We report educational efforts conducted in kindergarten, high school, university and working environment. For younger students, competitions take more the form of individual projects qualitatively evaluated by a jury, while at the university level they take the form of automatically graded homework addressing a research problem. The application domains can be very diverse: medicine, ecology, marketing, computer vision, recommendation, text processing, etc., and students enjoy being involved in creating challenges motivated by humanitarian purposes. Within a professional or working environment, competitions are used to help with upskilling, learning new approaches, and tools that can directly be used in one's job or to update one's resume to remain relevant and competitive in the market place.

Explainability and interpretability are two critical aspects of decision support systems. Despite their importance, it is only recently that researchers are starting to explore these aspects. This paper provides an introduction to explainability and interpretability in the context of apparent personality recognition. To the best of our knowledge, this is the first effort in this direction. We describe a challenge we organized on explainability in first impressions analysis from video. We analyze in detail the newly introduced data set, evaluation protocol, proposed solutions and summarize the results of the challenge. We investigate the issue of bias in detail. Finally, derived from our study, we outline research opportunities that we foresee will be relevant in this area in the near future.

Explainability and interpretability are two critical aspects of decision support systems. Within computer vision, they are critical in certain tasks related to human behavior analysis such as in health care applications. Despite their importance, it is only recently that researchers are starting to explore these aspects. This paper provides an introduction to explainability and interpretability in the context of computer vision with an emphasis on looking at people tasks. Specifically, we review and study those mechanisms in the context of first impressions analysis. To the best of our knowledge, this is the first effort in this direction. Additionally, we describe a challenge we organized on explainability in first impressions analysis from video. We analyze in detail the newly introduced data set, the evaluation protocol, and summarize the results of the challenge. Finally, derived from our study, we outline research opportunities that we foresee will be decisive in the near future for the development of the explainable computer vision field.