Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
167 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Can Probabilistic Feedback Drive User Impacts in Online Platforms? (2401.05304v2)

Published 10 Jan 2024 in cs.LG and cs.CY

Abstract: A common explanation for negative user impacts of content recommender systems is misalignment between the platform's objective and user welfare. In this work, we show that misalignment in the platform's objective is not the only potential cause of unintended impacts on users: even when the platform's objective is fully aligned with user welfare, the platform's learning algorithm can induce negative downstream impacts on users. The source of these user impacts is that different pieces of content may generate observable user reactions (feedback information) at different rates; these feedback rates may correlate with content properties, such as controversiality or demographic similarity of the creator, that affect the user experience. Since differences in feedback rates can impact how often the learning algorithm engages with different content, the learning algorithm may inadvertently promote content with certain such properties. Using the multi-armed bandit framework with probabilistic feedback, we examine the relationship between feedback rates and a learning algorithm's engagement with individual arms for different no-regret algorithms. We prove that no-regret algorithms can exhibit a wide range of dependencies: if the feedback rate of an arm increases, some no-regret algorithms engage with the arm more, some no-regret algorithms engage with the arm less, and other no-regret algorithms engage with the arm approximately the same number of times. From a platform design perspective, our results highlight the importance of looking beyond regret when measuring an algorithm's performance, and assessing the nature of a learning algorithm's engagement with different types of content as well as their resulting downstream impacts.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (44)
  1. Do recommender systems manipulate consumer preferences? A study of anchoring effects. Information Systems Research, 24(4):956–975, 2013.
  2. Automating automaticity: How the context of human choice affects the extent of algorithmic bias. Technical report, National Bureau of Economic Research, 2023.
  3. The welfare effects of social media. American Economic Review, 110(3):629–676, 2020.
  4. Online learning with feedback graphs: Beyond bandits. In Peter Grünwald, Elad Hazan, and Satyen Kale, editors, Proceedings of The 28th Conference on Learning Theory, COLT 2015, Paris, France, July 3-6, 2015, volume 40 of JMLR Workshop and Conference Proceedings, pages 23–35. JMLR.org, 2015.
  5. Finite-time analysis of the multiarmed bandit problem. Machine learning, 47:235–256, 2002a.
  6. The nonstochastic multiarmed bandit problem. SIAM journal on computing, 32(1):48–77, 2002b.
  7. A game-theoretic approach to recommendation systems with strategic content providers. In Advances in Neural Information Processing Systems (NeurIPS), pages 1118–1128, 2018.
  8. Content provider dynamics and coordination in recommendation ecosystems. In Advances in Neural Information Processing Systems (NeurIPS), 2020.
  9. Bandits meet mechanism design to combat clickbait in online recommendation. arXiv preprint arXiv:2311.15647, 2023.
  10. Understanding consumers’ social media engagement behaviour: An examination of the moderation effect of social media context. Journal of Business Research, 122:835–846, 2021.
  11. Estimating and penalizing induced preference shifts in recommender systems. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvári, Gang Niu, and Sivan Sabato, editors, International Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA, volume 162 of Proceedings of Machine Learning Research, pages 2686–2708. PMLR, 2022.
  12. The European Commission. Digital services act: EU’s landmark rules for online platforms enter into force. 2022.
  13. Online learning with dependent stochastic feedback graphs. In International Conference on Machine Learning, pages 2154–2163. PMLR, 2020.
  14. Preference dynamics under personalized recommendations. In David M. Pennock, Ilya Segal, and Sven Seuken, editors, EC ’22: The 23rd ACM Conference on Economics and Computation, Boulder, CO, USA, July 11 - 15, 2022, pages 795–816. ACM, 2022.
  15. Recommendations and user agency: the reachability of collaboratively-filtered information. In Conference on Fairness, Accountability, and Transparency (FAT* ’20), pages 436–445. ACM, 2020.
  16. Behaviorism is not enough: Better recommendations through listening to users. In Shilad Sen, Werner Geyer, Jill Freyne, and Pablo Castells, editors, Proceedings of the 10th ACM Conference on Recommender Systems, Boston, MA, USA, September 15-19, 2016, pages 221–224. ACM, 2016.
  17. Learning on the edge: Online learning with stochastic feedback graphs. Advances in Neural Information Processing Systems, 35:34776–34788, 2022.
  18. PAC bounds for multi-armed bandit and Markov decision processes. In International Conference on Computational Learning Theory, pages 255–270, 2002.
  19. Measuring emotional contagion in social media. PloS one, 10(11):e0142390, 2015.
  20. Filter Bubbles, Echo Chambers, and Online News Consumption. Public Opinion Quarterly, 80:298–320, 2016.
  21. Online learning with probabilistic feedback. In ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 4183–4187. IEEE, 2022.
  22. The stereotyping problem in collaboratively filtered recommender systems. In ACM Conference on Equity and Access in Algorithms (EAAMO), pages 6:1–6:10, 2021.
  23. Better algorithms for stochastic bandits with adversarial corruptions. In Conference on Learning Theory, pages 1562–1578. PMLR, 2019.
  24. Learning in stackelberg games with non-myopic agents. In Proceedings of the 23rd ACM Conference on Economics and Computation, pages 917–918, 2022.
  25. Risk aversion in learning algorithms and an application to recommendation systems. arXiv preprint arXiv:2205.04619, 2022.
  26. Elad Hazan et al. Introduction to online convex optimization. Foundations and Trends® in Optimization, 2(3-4):157–325, 2016.
  27. Beyond ads: Sequential decision-making algorithms in law and public policy. In Proceedings of the 2022 Symposium on Computer Science and Law, pages 87–100, 2022.
  28. Modeling content creator incentives on algorithm-curated platforms. CoRR, abs/2206.13102, 2022.
  29. Supply-side equilibria in recommender systems. CoRR, abs/2206.13489, 2022.
  30. Accurately interpreting clickthrough data as implicit feedback. 2005.
  31. Unbiased learning-to-rank with biased feedback. In Proceedings of the tenth ACM international conference on web search and data mining, pages 781–789, 2017.
  32. Fairness in learning: Classic and contextual bandits. Advances in neural information processing systems, 29, 2016.
  33. The challenge of understanding what users want: Inconsistent preferences and engagement optimization. In David M. Pennock, Ilya Segal, and Sven Seuken, editors, EC ’22: The 23rd ACM Conference on Economics and Computation, Boulder, CO, USA, July 11 - 15, 2022, page 29. ACM, 2022.
  34. Experimental evidence of massive-scale emotional contagion through social networks. Proceedings of the National academy of Sciences of the United States of America, 111(24):8788, 2014.
  35. Stochastic online learning with probabilistic graph feedback. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 4675–4682, 2020.
  36. Is a picture worth a thousand words? an empirical study of image content and social media engagement. Journal of Marketing Research, 57(1):1–19, 2020.
  37. Stochastic bandits robust to adversarial corruptions. In Proceedings of the 50th Annual ACM SIGACT Symposium on Theory of Computing, pages 114–122, 2018.
  38. From optimizing engagement to measuring value. In Madeleine Clare Elish, William Isaac, and Richard S. Zemel, editors, FAccT ’21: 2021 ACM Conference on Fairness, Accountability, and Transparency, Virtual Event / Toronto, Canada, March 3-10, 2021, pages 714–722. ACM, 2021.
  39. Auditing radicalization pathways on youtube. In Proceedings of the 2020 conference on fairness, accountability, and transparency, pages 131–141, 2020.
  40. Kevin Roose. The making of a youtube radical. The New York Times, 2019.
  41. Shaping feedback data in recommender systems with interventions based on information foraging theory. In Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining, pages 546–554, 2019.
  42. What are you optimizing for? aligning recommender systems with human values. CoRR, abs/2107.10939, 2021.
  43. Building human values into recommender systems: An interdisciplinary synthesis. CoRR, abs/2207.10192, 2022.
  44. Facebook knows instagram is toxic for teen girls, company documents show. The Wall Street Journal, 2021.
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets