Algorithms based on machine learning techniques are increasingly used for decision making in many different social contexts. For instance, algorithms screen job applicants, they recommend medical treatment, and they shape people`s online experiences through personalisation systems.
While such decision tools have the potential to enhance precision and cut costs by supporting or even replacing human decision makers, it is by no means a free lunch. For due to their architecture, the relevant class of algorithms make decisions in ways that are highly opaque to the human mind.
This raises a thorny trade-off between accuracy and explainability that we must learn to navigate. The 1st Aarhus Conference on Algorithmic Decision Making explores this trade-off, as well as other philosophical and practical questions that arise in the pursuit of the values of algorithmic decision making.

Speakers:

• Emily Sullivan
• Annette Zimmermann
• Chad Lee-Stronach
• Rune Nyrup
• Brent Mittelstadt
• Atoosa Kasirzadeh
• Frej Klem Thomsen
• Jens Christian Bjerring