Algorithmic Responsibility

Research in algorithmic responsibility at Spotify combines machine learning research with social science to ensure high quality data decisions and equitable algorithmic outcomes. We carry out in-depth research, perform product-focused case studies, as well as develop practical tools that teams can actually use. We do this by taking deep-dives into the characteristics of large-scale datasets, including understanding which creators are less accessible in both existing recommendations and new modalities such as voice, and developing machine learning approaches that combine multiple stakeholder objectives.

Latest Algorithmic Responsibility Publications

June 2019 | ICWSM

Local Trends in Global Music Streaming

Samuel F. Way, Jean Garcia Gathright, Henriette Cramer

April 2019 | CHI extended abstracts

Translation, Tracks & Data: an Algorithmic Bias Effort in Practice

Henriette Cramer, Jean Garcia-Gathright, Sravana Reddy, Avriel Epps, Aaron Springer, Romain Takeo Bouyer

October 2018 | CIKM

Towards a Fair Marketplace: Counterfactual Evaluation of the trade-off between Relevance, Fairness & Satisfaction in Recommendation Systems

Rishabh Mehrotra, James McInerney, Hugues Bouchard, Mounia Lalmas, Fernando Diaz