Fatigue-aware Bandits for Dependent Click Models

• URL: http://arxiv.org/abs/2008.09733v1
• Date: Sat, 22 Aug 2020 02:18:15 GMT
• Title: Fatigue-aware Bandits for Dependent Click Models
• Authors: Junyu Cao, Wei Sun, Zuo-Jun (Max) Shen, Markus Ettl
• Abstract summary: We consider an online learning setting where a platform learns a policy to recommend content that takes user fatigue into account. For each piece of content, its attractiveness to a user depends on its intrinsic relevance and a discount factor which measures how many similar contents have been shown. Based on user's feedback, the platform learns the relevance of the underlying content as well as the discounting effect due to content fatigue.
• Score: 11.887684896043883
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: As recommender systems send a massive amount of content to keep users engaged, users may experience fatigue which is contributed by 1) an overexposure to irrelevant content, 2) boredom from seeing too many similar recommendations. To address this problem, we consider an online learning setting where a platform learns a policy to recommend content that takes user fatigue into account. We propose an extension of the Dependent Click Model (DCM) to describe users' behavior. We stipulate that for each piece of content, its attractiveness to a user depends on its intrinsic relevance and a discount factor which measures how many similar contents have been shown. Users view the recommended content sequentially and click on the ones that they find attractive. Users may leave the platform at any time, and the probability of exiting is higher when they do not like the content. Based on user's feedback, the platform learns the relevance of the underlying content as well as the discounting effect due to content fatigue. We refer to this learning task as "fatigue-aware DCM Bandit" problem. We consider two learning scenarios depending on whether the discounting effect is known. For each scenario, we propose a learning algorithm which simultaneously explores and exploits, and characterize its regret bound.

Related papers

• Unveiling User Satisfaction and Creator Productivity Trade-Offs in Recommendation Platforms [68.51708490104687]
  We show that a purely relevance-driven policy with low exploration strength boosts short-term user satisfaction but undermines the long-term richness of the content pool. Our findings reveal a fundamental trade-off between immediate user satisfaction and overall content production on platforms.
  arXiv  Detail & Related papers  (2024-10-31T07:19:22Z)
• Uncovering the Interaction Equation: Quantifying the Effect of User Interactions on Social Media Homepage Recommendations [0.5030361857850012]
  We study how prior user interactions influence the content presented on users' homepage feeds across three major platforms: YouTube, Reddit, and X (formerly Twitter) We use a series of carefully designed experiments to gather data capable of uncovering the influence of specific user interactions on homepage content. This study provides insights into the behaviors of the content curation algorithms used by each platform, how they respond to user interactions, and also uncovers evidence of deprioritization of specific topics.
  arXiv  Detail & Related papers  (2024-07-09T20:47:34Z)
• System-2 Recommenders: Disentangling Utility and Engagement in Recommendation Systems via Temporal Point-Processes [80.97898201876592]
  We propose a generative model in which past content interactions impact the arrival rates of users based on a self-exciting Hawkes process. We show analytically that given samples it is possible to disentangle System-1 and System-2 and allow content optimization based on user utility.
  arXiv  Detail & Related papers  (2024-05-29T18:19:37Z)
• Measuring Strategization in Recommendation: Users Adapt Their Behavior to Shape Future Content [66.71102704873185]
  We test for user strategization by conducting a lab experiment and survey. We find strong evidence of strategization across outcome metrics, including participants' dwell time and use of "likes" Our findings suggest that platforms cannot ignore the effect of their algorithms on user behavior.
  arXiv  Detail & Related papers  (2024-05-09T07:36:08Z)
• User Welfare Optimization in Recommender Systems with Competing Content Creators [65.25721571688369]
  In this study, we perform system-side user welfare optimization under a competitive game setting among content creators. We propose an algorithmic solution for the platform, which dynamically computes a sequence of weights for each user based on their satisfaction of the recommended content. These weights are then utilized to design mechanisms that adjust the recommendation policy or the post-recommendation rewards, thereby influencing creators' content production strategies.
  arXiv  Detail & Related papers  (2024-04-28T21:09:52Z)
• Incentivizing High-Quality Content in Online Recommender Systems [80.19930280144123]
  We study the game between producers and analyze the content created at equilibrium. We show that standard online learning algorithms, such as Hedge and EXP3, unfortunately incentivize producers to create low-quality content.
  arXiv  Detail & Related papers  (2023-06-13T00:55:10Z)
• How Bad is Top-$K$ Recommendation under Competing Content Creators? [43.2268992294178]
  We study the user welfare guarantee through the lens of Price of Anarchy. We show that the fraction of user welfare loss due to creator competition is always upper bounded by a small constant depending on $K$ and randomness in user decisions.
  arXiv  Detail & Related papers  (2023-02-03T19:37:35Z)
• Reward Shaping for User Satisfaction in a REINFORCE Recommender [24.65853598093849]
  We propose a policy network and a satisfaction imputation network to learn which actions are satisfying to the user. The role of the imputation network is to learn which actions are satisfying to the user; while the policy network, built on top of REINFORCE, decides which items to recommend.
  arXiv  Detail & Related papers  (2022-09-30T01:29:12Z)
• An Empirical Investigation of Personalization Factors on TikTok [77.34726150561087]
  Despite the importance of TikTok's algorithm to the platform's success and content distribution, little work has been done on the empirical analysis of the algorithm. Using a sock-puppet audit methodology with a custom algorithm developed by us, we tested and analysed the effect of the language and location used to access TikTok. We identify that the follow-feature has the strongest influence, followed by the like-feature and video view rate.
  arXiv  Detail & Related papers  (2022-01-28T17:40:00Z)
• CLUE: Contextualised Unified Explainable Learning of User Engagement in Video Lectures [6.25256391074865]
  We propose a new unified model, CLUE, which learns from the features extracted from public online teaching videos. Our model exploits various multi-modal features to model the complexity of language, context information, textual emotion of the delivered content.
  arXiv  Detail & Related papers  (2022-01-14T19:51:06Z)
• Dynamic Slate Recommendation with Gated Recurrent Units and Thompson Sampling [6.312395952874578]
  We consider the problem of recommending relevant content to users of an internet platform in the form of lists of items, called slates. We introduce a variational Bayesian Recurrent Neural Net recommender system that acts on time series of interactions between the internet platform and the user. We show experimentally that explorative recommender strategies perform on par or above their greedy counterparts.
  arXiv  Detail & Related papers  (2021-04-30T15:16:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.