Consequential Ranking Algorithms and Long-term Welfare

Abstract

Ranking models are typically designed to provide rankings that optimize some measure of immediate utility to the users. As a result, they have been unable to anticipate an increasing number of undesirable long-term consequences of their proposed rankings, from fueling the spread of misinformation and increasing polarization to degrading social discourse. Can we design ranking models that understand the consequences of their proposed rankings and, more importantly, are able to avoid the undesirable ones? In this paper, we first introduce a joint representation of rankings and user dynamics using Markov decision processes. Then, we show that this representation greatly simplifies the construction of consequential ranking models that trade off the immediate utility and the long-term welfare. In particular, we can obtain optimal consequential rankings just by applying weighted sampling on the rankings provided by models that maximize measures of immediate utility. However, in practice, such a strategy may be inefficient and impractical, specially in high dimensional scenarios. To overcome this, we introduce an efficient gradient-based algorithm to learn parameterized consequential ranking models that effectively approximate optimal ones. We showcase our methodology using synthetic and real data gathered from Reddit and show that ranking models derived using our methodology provide ranks that may mitigate the spread of misinformation and improve the civility of online discussions.

References in corpus (13)

• Fairness of Exposure in Rankings
• Anyone Can Become a Troll: Causes of Trolling Behavior in Online Discussions
• FA*IR: A Fair Top-k Ranking Algorithm
• Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Review
• Optimal control as a graphical model inference problem
• Designing Fair Ranking Schemes
• How Visibility and Divided Attention Constrain Social Contagion
• A Short-term Intervention for Long-term Fairness in the Labor Market
• Learning and Forecasting Opinion Dynamics in Social Networks
• Deconvolving Feedback Loops in Recommender Systems
• Policy Learning for Fairness in Ranking
• VIP: Incorporating Human Cognitive Biases in a Probabilistic Model of Retweeting
• Choice by Elimination via Deep Neural Networks

Cited by in corpus (3)

• Controlling Fairness and Bias in Dynamic Learning-to-Rank
• Optimal Decision Making Under Strategic Behavior
• Fairness with Continuous Optimal Transport