Scheduling with Uncertain Holding Costs and its Application to Content Moderation
Caner Gocmen, Thodoris Lykouris, Deeksha Sinha, Wentao Weng

TL;DR
This paper introduces a novel index-based algorithm, OaRC, for scheduling in systems with uncertain, state-dependent holding costs, demonstrating its asymptotic optimality and superior performance in content moderation scenarios.
Contribution
The paper develops the Opportunity-adjusted Remaining Cost (OaRC) algorithm, addressing the limitations of traditional rules under uncertain holding costs, and proves its asymptotic optimality with a scalable regret bound.
Findings
OaRC outperforms existing scheduling policies in simulations.
Regret of OaRC scales as old^{1.5}\u221a{N}, independent of state-space size.
OaRC achieves asymptotic optimality as system size N grows.
Abstract
In content moderation for social media platforms, the cost of delaying the review of a content is proportional to its view trajectory, which fluctuates and is apriori unknown. Motivated by such uncertain holding costs, we consider a queueing model where job states evolve based on a Markov chain with state-dependent instantaneous holding costs. We demonstrate that in the presence of such uncertain holding costs, the two canonical algorithmic principles, instantaneous-cost (-rule) and expected-remaining-cost (-rule), are suboptimal. By viewing each job as a Markovian ski-rental problem, we develop a new index-based algorithm, Opportunity-adjusted Remaining Cost (OaRC), that adjusts to the opportunity of serving jobs in the future when uncertainty partly resolves. We show that the regret of OaRC scales as , where is the maximum length of a…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
Scheduling with Time-Evolving Uncertainty for Content Review Prioritization in Social Media· youtube
Taxonomy
TopicsHate Speech and Cyberbullying Detection · Spam and Phishing Detection
