Recommender Systems Optimization Goals

In part two of the Data Skeptic Recommender Systems season finale, Kyle asks a deceptively difficult question: what should recommender systems actually optimize for? Drawing on conversations from across the season, the episode explores engagement, filter bubbles, popularity bias, fairness, human curation, embeddings, and the growing role—and risks—of large language models in shaping what gets recommended to us.

Guests

Kyle Polich: Kyle is the founder of Data Skeptic, a popular podcast about artificial intelligence, machine learning, and data science. Outside of hosting the show, he runs a boutique consulting group that helps small and medium enterprise companies deploy data driven automated solutions in the cloud.

Gregor Donabauer: Since October 2021, I have been a PhD student and research officer at the Chair of Information Science at the University of Regensburg (Germany). I previously also worked as a research assistant at the University of Milano-Bicocca (Italy) in Prof. Dimitri Ognibene’s BiConnect Lab, collaborating on the VolkswagenStiftung-funded COURAGE research project. I hold both a Bachelor’s degree in Information Science/Business Information Systems and a Master’s degree in Information Science from the University of Regensburg. My research interests focus on Natural Language Processing and Graph Machine Learning, particularly in the areas of Information Retrieval and applications in the medical domain. In addition, I serve as an elected committee member of the British Computer Society’s Information Retrieval specialist Group (BCS IRSG) for the 2023–2025 term.

Andreea Iana: I am a postdoctoral researcher in the Data and Web Science Group at the University of Mannheim. I hold a PhD in Recommender Systems from the University of Mannheim, where I was advised advised by Prof. Heiko Paulheim (University of Mannheim) and Prof. Goran Glavaš (University of Würzburg). During my PhD, I was a visting researcher in the WüNLP Group at the University of Würzburg. Before that, I completed my M.Sc. in Business Informatics at the University of Mannheim, and my B.Sc. in Liberal Arts and Sciences from the University College Maastricht, in the Netherlands. I work on advancing responsible and inclusive AI for information access, with a particular focus on multilinguality, cross-lingual IR, and investigating and mitigating algorithmic bias in retrieval and recommendation systems.

Hieu Le: I am currently serving as a Senior Technologist at the Federal Trade Commission’s Office of Technology, where I leverage my expertise on privacy, automated systems, and applied machine learning to further the agency’s mission to protect consumers and promote fair competition. Previously, I was a Postdoctoral Research Fellow within Roya Ensafi’s Censored Planet Lab in the Computer Science and Engineering (CSE) School at the University of Michigan, Ann Arbor. I collaborated on projects that examined the impact of geoblocking on citizens in sanctioned states as well as the development of machine learning and outlier detection methodologies to identify censorship events and content. I received my PhD in Electrical Engineering and Computer Science (EECS) at UC Irvine and was advised by Athina Markopoulou and part of the ProperData NSF Frontier project. My doctoral studies centers on privacy and networking to improve the (1) transparency of data collection practices of different platforms such as the web, smart TVs and Oculus VR; and the (2) control of data collection by designing frameworks and methodologies to automate the pain-points of privacy-enhancing technologies (PETs) that block advertising and tracking on the web.

Boya Xu: Boya Xu is an Assistant Professor of Marketing at the Pamplin College of Business, Virginia Tech. Boya holds a Ph.D. in Marketing and an M.A. in Economics from Duke University, and a B.S. in Statistics from Zhejiang University, China. Boya conducts research in quantitative marketing, focusing on platform designs, content marketing, and the economies of emerging technologies. Her research combines methods from econometrics, online experimentation, and unstructured data analytics. Boya was the winner of the 2024 Doctoral Dissertation Research Award at the American Statistics Association (Marketing Section) and a recipient of the 2023 NET Institute Research Grant.

Václav Blahut: Václav mastered AI & NLP course at FI MUNI and is now teaching machines how to recommend articles and other content to Seznam.cz users. He's been doing it for more than six years now. Currently, he is trying to figure out how to improve content-based recommendation of external content with sparse feedback and how to improve recommendations in general.

Anas Buhayh: Anas Buhayh is a PhD student at the Recommender Systems Lab at the University of Colorado Boulder, where his research investigates alternative architectures and technologies for building responsible and governable AI systems. His work emphasizes the importance of addressing the needs of stakeholders in sociotechnical systems. Anas also volunteers as a consultant for recommender system design and development. A Fulbright alumnus from Libya, Anas holds a master’s degree in Information Management from Central Michigan University and a bachelor’s degree in Automation from the College of Computer Technologies in Tripoli, Libya. His professional experience includes roles in monitoring and evaluation with the World Health Organization and the International Rescue Committee, where he focused on aligning project outcomes with stakeholder needs. Anas is passionate about data analytics, machine learning, and systems design. In his free time, he enjoys programming, playing the guitar, and staying active.

Rebecca Salganik: Rebecca Salganik is a computer scientist and musician whose research bridges artificial intelligence, fairness, and music discovery. She is currently pursuing a PhD in Computer Science at the University of Rochester in New York under Prof. Jian Kang, focusing on Individual Fairness in Graphs and Discovery in Recommender Systems. She completed her M.Sc. at Université de Montréal (MILA), where her thesis on fairness-aware music recommendation explored how to mitigate popularity bias in algorithmic systems—read it here. Her research, published at top conferences including ECIR, KDD, FAccT, and RecSys, examines how Graph Neural Networks, Fairness, and Music Recommender Systems can be designed to amplify diverse voices and create more equitable pathways for artistic expression. Rebecca has held research positions at Pinterest, Pandora Labs, Lyft, and MIT Lincoln Labs, and she organizes DEFirst, a reading group focused on fairness in information retrieval. Her background as a classically trained vocalist from McGill University deeply informs her interdisciplinary approach to building ethical AI systems that respect the nuances of musical expression.

David Liu: David Liu is an Assistant Research Professor at Cornell University's Center for Data Science for Enterprise and Society where he examines how AI impacts society. Specifically, he is interested in 1) understanding how machine learning models homogenize heterogeneous populations and 2) building models that better capture the unique preferences and identities of marginalized individuals. Prior to Cornell, David obtained a Ph.D. in Computer Science from Northeastern University, with the support of the NSF GRFP, and a B.S.E. from Princeton University. He has worked in industry as a research-scientist intern at Meta, a sociotechnical researcher at Taraaz, and a software engineer at Bloomberg LP.

Cory Zechmann: a content curator working in streaming television with 16 years of experience running the music blog "A Silence No Good."

Ekaterina (Kat) Fedorova: About Hello! I’m Ekaterina (or Kat), an NSF CSGrad4US fellow and an EECS PhD student at MIT. At MIT, I currently work Dr. Chara Podimata and Dr. Costis Daskalakis. I’m interested in a combination of theoretical (CS and micro econ) algorithmic game theory topics like strategic classification, no-regret algorithms, and learning more broadly. More concretely, I’m currently working on projects that involve Correlated agent strategies in strategic learning settings Allocation mechanisms to information-disparate strategic agents Learning from biased data and truncated statistics I spent the first year of my PhD at UPenn CS (where I also worked closely with Dr. Danaë Metaxa) before moving to MIT EECS to continue my work. Before my time as a PhD student, I worked as a full time researcher for Dr. Anna Costello at UChicago. I received a B.A. from UC Berkeley where I studied statistics and economics. In my free time, I like to try to spot opossums around my campus and try new teas

Ashmi Banerjee: Ashmi is currently a doctoral researcher at the Chair of Connected Mobility at the Technical University of Munich. Her research focuses on Recommender Systems, specifically in the tourism domain. She graduated with a master's degree in Computer Science in 2019 from the same university and also holds three years of industry experience at different companies across Germany. She is passionate about using technology to automate tedious tasks and is always excited to tackle new technical challenges. Over the past two years, she has delivered over 50 tech talks across 15+ countries on three continents. She was named one of the 100 technologists to watch for 2023 and won the Google Developer Expert Community Award (Rising Star), the 2023 Women Who Code Applaud Her Award (Data Science), and the DevelopHER Awards 2022 (Emerging Talent). As a Google Women Techmakers (WTM) Ambassador and diversity advocate, she is dedicated to closing the gender gap in STEM through her involvement in various women in STEM networks. When not sitting in front of her computer, she can usually be found traveling, collecting passport stamps, and fridge magnets. ✈️

Ervin Dervishaj: Applied Scientist intern Amazon and PhD fellow in Recommender Systems @ University of Copenhagen

Yashar Deldjoo: Dr. Yashar Deldjoo is a senior research scientist and Associate Professor of Computer Science at the Polytechnic University of Bari, Italy. His research agenda focuses on integrating trustworthy and responsible AI into modern recommender systems, particularly those leveraging Generative AI and multi‑agent large language models. Over the past several years, he has played a pivotal role in understanding, categorizing and mitigating emerging risks in these systems—including hallucinations, biases, reasoning challenges, content‑safety concerns, adversarial resilience and goal misalignment—and in developing holistic evaluation frameworks to ensure safe deployment. Dr. Deldjoo recently led a multidisciplinary collaboration to author Next‑Generation Recommender Systems under Generative AI (Gen‑RecSys), a forthcoming book within the Foundations and Trends® in Information Retrieval series (FNT&IR 2025), which explores advanced generative techniques for recommender systems. Beyond research, Dr. Deldjoo provides extensive service to the academic community. He is an Associate Editor for IEEE Transactions on Knowledge and Data Engineering (TKDE), ACM Computing Surveys (CSUR), and ACM Transaction on Recommender Systems (TORS). He has served as a Track Chair or Senior Program Committee member at major IR and AI conferences, including KDD, The Web Conference (WWW), CIKM, SIGIR, RecSys, and ECAI, among others.

Hannes Rosenbusch: Hannes Rosenbusch lurks in the cafés of Amsterdam, writing obscure novels about the afterlife. To advance his scientific research, he frequently lures psychology students into the laboratories beneath Roeterseiland Campus. Most of his colleagues question the validity of his experiments and publications. He has a German accent and caffeine-induced muscle spasms.