I am a Research Scientist in the Personalization Economics group at Spotify. My research focuses on data-driven decision-making and digitalization in marketing, health, and the public sector. More broadly, I am interested in developing and applying statistical and econometric methods based on causal inference, machine learning, and decision theory to ensure that digital technologies and data-driven decisions are effective, robust, fair, and explainable for use in practice. My work to date relates to two themes:
Applications of causal machine learning in digitalization, including machine learning of contextual, dynamic and personalized decision policies in digital marketing and health, and addressing group bias in causal machine learning of treatment effects on digital platforms.
The value of digitalization, including the benefit of large-scale observational data from digital technologies for prediction and policy evaluation, and how models trained on such data can outperform human experts in complex decision-making problems.
Please see my research page for details.
I hold a PhD from the Department of Management, Technology, and Economics at ETH Zurich under the supervision of Stefan Feuerriegel and Florian von Wangenheim. During my PhD, I visited Stanford Graduate School of Business and interned as Machine Learning Research Scientist at Booking.com. I also joined the non-profit organization Algorithm Audit as a contributor on statistical methodology, which I continue to be involved in. Previously, I completed double bachelors and master's degrees in Statistics and Business & Economics at Lund University in Sweden and worked in market research and digital advertising. Details are available on my experience page.
News
2 July 2024: Our session proposal "Advances in Causal Machine Learning" with Ravi B. Sojitra and Apoorva Lal moderated by Rose Tan has been accepted for presentation at the NABE Tech Economics Conference (TEC 2024) in Seattle, Oct 27-29!
15 May 2024: I represented Spotify Research on the industry career panel at the 2024 American Causal Inference Conference in Seattle.
17 Jan 2024: Our working paper on algorithmic bias in treatment effect estimation has been accepted for presentation at the Workshop on Platform Analytics (WoPA) 2024 in San Diego.
29 Nov 2023: In Jan 2024, I will be joining Spotify as a Research Scientist.
16 Nov 2023: I have successfully defended my PhD thesis, titled: "Causal Machine Learning for Data-Driven Decision-Making: Methods and Applications"
28 Oct 2023: I have been invited to present our work "Off-Policy Learning of Content Promotions: Optimizing Digital Distribution Channels" at WAN-IFRA, World Association of News Publishers, on Nov 21
26 Oct 2023: I will chair the session "Machine Learning and Optimization Applications" at the 2023 INFORMS Annual Meeting where I will also present our work "Off-Policy Learning of Content Promotions: Optimizing Digital Distribution Channels"
26 May 2023: I will be presenting "Off-Policy Learning of Content Promotions: Optimizing Digital Distribution Channels" at the QME conference 2023, EURO Workshop 2023, and INFORMS Revenue Management and Pricing Section Conference 2023, and as a poster at EC'23.
1 May 2023: "Off-Policy Learning of Content Promotions: Optimizing Digital Distribution Channels" was accepted for presentation at the MSOM2023 conference in Montreal, Canada, June 25-26.
30 March 2023: I will be presenting our work "Detecting and Mitigating Discriminatory Bias in Uplift Modeling: A Causal Fairness Approach with a Field Experiment" at the Theory+Practice in Marketing Conference 2023 in Lausanne, May 29-31.
20 March 2023: This April to June, I will be visiting the research group of Jann Spiess, Stefan Wager, and Lihua Lei at Stanford GSB, Department of Economics and the Operations, Information & Technology area.
21 Feb 2023: Happy to share that "Detecting and Mitigating Discriminatory Bias in Uplift Modeling: A Causal Fairness Approach with a Field Experiment" has been accepted for presentation at the ISMS Marketing Science Conference 2023 in Miami, June 8-10.
15 Feb 2023: I will be presenting "Detecting and Mitigating Discriminatory Bias in Uplift Modeling: A Causal Fairness Approach with a Field Experiment" at the American Causal Inference Conference 2023 in Austin, May 24-26.
31 Jan 2023: Our work-in-progress "Detecting and Mitigating Discriminatory Bias in Uplift Modeling: A Causal Fairness Approach with a Field Experiment" has been accepted for presentation at the Marketing Science conference on Diversity, Equity and Inclusion at SMU Cox School of Business in Dallas.
24 Jan 2023: My nomination for the Doctoral Consortium at the ISMS Marketing Science Conference 2023 in Miami was accepted.
Dec 4 2022: The submission of Algorithm Audit to the AI Audit challenge at Stanford HAI and Stanford Cyber Policy center made it to the final round!
30 Nov 2022: I have finished my internship at Booking.com in the Machine Learning Foundations, Core Data Science, and Algorithmic Fairness Task Force groups. I can very much recommend research internships to other doctoral students.
24 Jun 2022: Our working paper "Learning Optimal Dynamic Treatment Regimes Using Causal Tree Methods in Medicine " is accepted for the Machine Learning in Healthcare conference 2022 at Duke University.
28 Feb 2022: I will join the Machine Learning Foundations and Core Data Science groups at Booking.com for a PhD research internship at their headquarters in Amsterdam later this year.