Many multi-user scenarios are characterised by a combinatorial nature, i.e., an algorithm can take meaningful decisions for the users only if all their requirements and preferences are considered at the same time to select a solution from a huge potential space of possible system decisions. Sharing economy application, where users aim to find peers to form teams with in order to accomplish a task, and situations in which a limited number of potentially different resources, e.g. hotel rooms, must be distributed to users who have preferences over them are examples of such scenarios.
Continue reading How hard is to be fair in multi-user combinatorial scenarios?
On February 3rd a group of twenty five stakeholders joined us at the Digital Catapult in London for our first discussion workshop.
The User Engagement workpackage of the project focuses on gathering together professionals from industry, academia, education, NGOs and research institutes in order to discuss societal and ethical issues surrounding the design, development and use of algorithms on the internet. We aim to create a space where these stakeholders can come together and discuss their various concerns and perspectives. This includes finding differences of opinion. For example, participants from industry often view algorithms as proprietary and commercially sensitive whereas those from NGOs frequently call for greater transparency in algorithmic design. It is important for us to draw out these kinds of varying perspectives and understand in detail the reasoning that lies behind them. Then, combined with the outcomes of the other project workpackages, we can identify points of resolution and produce outputs that seek to advance responsibility on algorithm driven internet platforms.
Continue reading First UnBias Multi-Stakeholder Workshop
An important topic considered this year at the International Conference on Neural Information Processing Systems (NIPS), one of the prime outlets for machine learning and Artificial Intelligence research in the world, is the connection between machine learning, law and ethics. In particular, a paper presented by Moritz Hardt, Eric Price, and Nathan Srebro focused on Equality of Opportunity in Supervised Learning.
Continue reading Fair machine learning techniques and the problem of transparency
A lot has been said about algorithms working as gatekeepers and making decisions on our behalf, often without us noticing it. I can surely find an example in my daily life where I do notice it and benefit from it. This happens when I use the “Discover Weekly” Spotify play-list. By comparing my listening habits to that of other users with similar but not identical choices, Spotify allows information on the fringes to be shared. It is thus “tailored” to my music taste, and it is incredibly accurate in predicting things I would like. Besides, it lets me discover new music and bands and in many occasions can also take me back in time with some tunes I have probably not listened to for a long time.
Continue reading Algorithmic discrimination: are you IN or OUT?