Session 3: Bias and Fairness 2

Date: Tuesday October 15, 14:30 PM – 15:15 PM (GMT+2)
Room: Petruzzelli Theater
Session Chair: Elisabeth Lex

• RES 🕓15Biased User History Synthesis for Personalized Long-Tail Item Recommendation
  by Keshav Balasubramanian (University of Southern California), Abdulla Alshabanah (University of Southern California), Elan Markowitz (University of Southern California), Greg Ver Steeg (University of California Riverside) and Murali Annavaram (University of Southern California)

  Recommendation systems connect users to items and create value chains in the internet economy. Recommendation systems learn from past user-item interaction histories. As such, items that have short interaction histories, either because they are new or not popular, have been shown to be disproportionately under-recommended. This long-tail item problem can exacerbate model bias, and reinforce poor recommendation of tail items. In this paper, we propose biased user history synthesis, to not only address this problem but also achieve better personalization in recommendation systems. As a result, we concurrently improve tail and head item recommendation performance. Our approach is built on a tail item biased User Interaction History (UIH) sampling strategy and a synthesis model that produces an augmented user representation from the sampled user history. We provide a theoretical justification for our approach using information theory and demonstrate through extensive experimentation, that our model outperforms state-of-the-art baselines on tail, head, and overall recommendation. The source code is available at https://github.com/lkp411/BiasedUserHistorySynthesis.

  Full text in ACM Digital Library

• RES 🕓15The Fault in Our Recommendations: On the Perils of Optimizing the Measurable
  by Omar Besbes (Columbia University), Yash Kanoria (Columbia University) and Akshit Kumar (Columbia University)

  Recommendation systems are widespread, and through customized recommendations, promise to match users with options they will like. To that end, data on engagement is collected and used. Most recommendation systems are ranking-based, where they rank and recommend items based on their predicted engagement. However, the engagement signals are often only a crude proxy for user utility, as data on the latter is rarely collected or available. This paper explores the following question: By optimizing for measurable proxies, are recommendation systems at risk of significantly under-delivering on user utility? If that is indeed the case, how can one improve utility which is seldom measured?To study these questions, we introduce a model of repeated user consumption in which, at each interaction, users select between an outside option and the best option from a recommendation set. Our model accounts for user heterogeneity, with the majority preferring “popular” content, and a minority favoring “niche” content. The system initially lacks knowledge of individual user preferences but can learn these preferences through observations of users’ choices over time. Our theoretical and numerical analysis demonstrate that optimizing for engagement signals can lead to significant utility losses. Instead, we propose a utility-aware policy that initially recommends a mix of popular and niche content. We show that such a policy substantially improves utility despite not measuring it. As the platform becomes more forward-looking, our utility-aware policy achieves the best of both worlds: near-optimal user utility and near-optimal engagement simultaneously. Our study elucidates an important feature of recommendation systems; given the ability to suggest multiple items, one can perform significant exploration without incurring significant reductions in short term engagement. By recommending high-risk, high-reward items alongside popular items, systems can enhance discovery of high utility items without significantly affecting engagement.

  Full text in ACM Digital Library

• RES 🕓15Fair Reciprocal Recommendation in Matching Markets
  by Yoji Tomita (CyberAgent Inc.) and Tomohiko Yokoyama (The University of Tokyo)

  Recommender systems play an increasingly crucial role in shaping people’s opportunities, particularly in online dating platforms. It is essential from the user’s perspective to increase the probability of matching with a suitable partner while ensuring an appropriate level of fairness in the matching opportunities.

  We investigate reciprocal recommendation in two-sided matching markets between agents divided into two sides. In our model, a match is considered successful only when both individuals express interest in each other. Additionally, we assume that agents prefer to appear prominently in the recommendation lists presented to those on the other side. We define each agent’s opportunity to be recommended and introduce its fairness criterion, envy-freeness, from the perspective of fair division theory. The recommendations that approximately maximize the expected number of matches, empirically obtained by heuristic algorithms, are likely to result in significant unfairness of opportunity. Therefore, there can be a trade-off between maximizing the expected matches and ensuring fairness of opportunity. To address this challenge, we propose a method to find a policy that is close to being envy-free by leveraging the Nash social welfare function. Experiments on synthetic and real-world datasets demonstrate the effectiveness of our approach in achieving both relatively high expected matches and fairness for opportunities of both sides in reciprocal recommender systems.

  Full text in ACM Digital Library

Back to program