Algorithmic recourse: from theory to practice

Isabel Valera (Saarland University, Germany) will cover the concept of algorithmic recourse, showing some results and practical solutions.

In this talk Valera will introduce the concept of algorithmic recourse, which aims to help individuals affected by an unfavorable algorithmic decision to recover from it. First, she will show that while the concept of algorithmic recourse is strongly related to counterfactual explanations, existing methods for the later do not directly provide practical solutions for algorithmic recourse, as they do not account for the causal mechanisms governing the world. Then, she will show theoretical results that prove the need of complete causal knowledge to guarantee recourse and show how algorithmic recourse can be useful to provide novel fairness definitions that short the focus from the algorithm to the data distribution. Such novel definition of fairness allows us to distinguish between situations where unfairness can be better addressed by societal intervention, as opposed to changes on the classifiers. Finally, she will show practical solutions for (fairness in) algorithmic recourse, in realistic scenarios where the causal knowledge is only limited.

Teacher / speaker

Isabel Valera

Isabel Valera is a full Professor on Machine Learning at the Department of Computer Science of Saarland University in Saarbrücken (Germany), and Adjunct Faculty at MPI for Software Systems in Saarbrücken (Germany).
She is a fellow of the European Laboratory for Learning and Intelligent Systems (ELLIS), where she is part of the Robust Machine Learning Program and of the Saarbrücken Artificial Intelligence & Machine learning (Sam) Unit.
Prior to this, she was an independent group leader at the MPI for Intelligent Systems in Tübingen (Germany) until the end of the year. She has held a German Humboldt Post-Doctoral Fellowship, and a “Minerva fast track” fellowship from the Max Planck Society. She obtained her PhD in 2014 and MSc degree in 2012 from the University Carlos III in Madrid (Spain), and worked as postdoctoral researcher at the MPI for Software Systems (Germany) and at the University of Cambridge (UK).

Research interests

Valera's research focuses on developing machine learning methods that are flexible, robust, interpretable and fair. Flexible means they are capable of modeling complex real-world data, which are often heterogeneous in nature and present temporal dependencies. Secondly, she aims to improve the robustness of machine learning algorithms to outliers, missing data and mixed statistical data types. Finally, she works on making algorithms fairer and interpretable – if they are part of important decision-making processes, the outcomes should be fair and explainable.
Her research can be applied in a broad range of fields, from medicine and psychiatry to social and communication systems. Recently, she also began putting a special focus on consequential decision making in several domains, including hiring processes, pre-trial bail, or loan approval.

Sense & Sensibility of AI

AI has an increasing influence on our daily lives, examples include automated decision-making for high-stake decisions such as mortgages and loans, automated risk assessments for bail or recommenders on the internet. These AI systems carry the risk of creating filter bubbles and polarization. While AI is being rolled out into society, the discussion on how AI-based systems may align with and even affect our values, is pushed to the forefront. We gave the computer senses, but how can we give it sensibility? It requires a multi-disciplinary view, where both technical and non-technical perspectives have a prominent place.

In our lecture series ‘Sense & Sensibility of AI,’ we aim for Ph.D. Students to learn about the different aspects of Ethics in AI, not only to become aware of them but also to learn about the impact of AI on society and about methodologies to identify, assess, and possibly address ethical issues. The monthly seminars tackle subjects such as bias and fairness, privacy, trustworthiness, balancing technical, social, and regulatory perspectives.

The series is targeted towards doctoral students working in the broad field of AI and data science. To understand the lectures in full, it may be required to have a background in the technical aspects of AI/machine learning.

Would you like to receive an announcement of each seminar in your inbox? Register for our monthly reminder!

Sense & Sensibility of AI is a seminar series developed by Flemish AI Academy in collaboration and with the support of all our partners, all universities in Flanders, and Knowledge Center Data & Society.

More info & registration ⇗