ﻻ يوجد ملخص باللغة العربية
Domain name registrars and URL shortener service providers place advertisements on the parked domains (Internet domain names which are not in service) in order to generate profits. As the web contents have been removed, it is critical to make sure the displayed ads are directly related to the intents of the visitors who have been directed to the parked domains. Because of the missing contents in these domains, it is non-trivial to generate the keywords to describe the previous contents and therefore the users intents. In this paper we discuss the adaptive keywords extraction problem and introduce an algorithm based on the BM25F term weighting and linear multi-armed bandits. We built a prototype over a production domain registration system and evaluated it using crowdsourcing in multiple iterations. The prototype is compared with other popular methods and is shown to be more effective.
In this paper, we consider online learning in generalized linear contextual bandits where rewards are not immediately observed. Instead, rewards are available to the decision-maker only after some delay, which is unknown and stochastic. We study the
Frequently-Asked-Question (FAQ) retrieval provides an effective procedure for responding to users natural language based queries. Such platforms are becoming common in enterprise chatbots, product question answering, and preliminary technical support
We consider the linear contextual bandit problem with resource consumption, in addition to reward generation. In each round, the outcome of pulling an arm is a reward as well as a vector of resource consumptions. The expected values of these outcomes
We study contextual bandits with ancillary constraints on resources, which are common in real-world applications such as choosing ads or dynamic pricing of items. We design the first algorithm for solving these problems that handles constrained resou
We consider the problem of learning to choose actions using contextual information when provided with limited feedback in the form of relative pairwise comparisons. We study this problem in the dueling-bandits framework of Yue et al. (2009), which we