Algorithm Watchdog

As part of an ongoing initiative, supported by the CyCAT EU-funded project, we are establishing a watchdog activity that will bring together interdisciplinary experts
to produce recommendations for the oversight of algorithms used in application domains that are increasingly having profound socio economic impact.

We envision that this watchdog will initially operate as an informal forum to discuss current problems that arise from the use of algorithms, especially those that raise privacy, fairness, transparency, and accountability issues, but also where they are seen to impact markets, competition, democracy, and civil rights, and produce a series of reports that can be used to influence policy makers and industry as well as inform the general public. Our long-term goal is to establish a semi-formal group of experts who can identify key issues, challenges and can lay out potential solutions through specific recommendations to the concrete issues discussed on these workshops.

To gauge interest in this activity, we will be running a series of events over the next few months to discuss different timely topics related to algorithmic bias, discrimination, their effects and potential mitigation strategies. We will bring together an international and diverse group of experts in order to be able to provide a technological perspective which – we believe – is currently underrepresented in the discussions on the aforementioned topics. For this reason, at this time we are welcoming contributors who have expertise in related STEM areas.

Following each session we will convert the discussion into a short publication featuring key recommendations, to which all participants can contribute as authors.

Our first workshop will focus on Algorithmic Decision Making in Recruitment and HR, and will be held online on the 17th March 2021 between 16:00-18:00 GMT.

Such systems are being used in hiring more and more frequently for different reasons, such as to increase efficiency or to mitigate bias in the recruitment process. These decisions are present throughout the hiring funnel from the step of wording the job description and advert through ranking candidates for interviewing to assess video-interviews for particular personality traits of the applicants. We have conducted interviews with stakeholders from the field of recruitment, such as large-scale employers, recruitment firms and expert advocacy groups in order to gain insight for the identification of the main challenges and problems the use of algorithmic decision-making systems poses in hiring. This was followed by a short project where we utilized the help of graduate student participants of the FATE (Fairness, Accountability, Transparency and Ethics) CyCAT Winter School in order to design a draft methodology to monitor bias in a given recruitment platform. In the workshop we will share our findings and will then assess these results together with the group of experts leading to the design of an assessment methodology and requirements.

If you would like to participate, please use the following link to sign up by 20th February.
https://www.eventbrite.co.uk/e/136625162539/

If you would like to be added to our distribution list regarding upcoming events and news, please let us know by registering your interest on the following form: https://forms.gle/R6Z4CjUvRycDQnG56

If you have any questions please do not hesitate to contact Prof. Michael Rovatsos (michael.rovatsos@ed.ac.uk) or Dr Lena Podoletz (lena.podoletz@ed.ac.uk).

17 March 2021 4:00 pm (GMT)

Online

2 hours