Explore/Exploit Schemes for Web Content Optimization

By Deepak K. Agarwal, Bee-Chung Chen, Pradheep Elango. 2009

126 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Lihong Li
2. Nathan Kallus
3. J. Langford
4. Liang Tang
5. Tao Li
6. Thorsten Joachims
7. Lei Li
8. Chunqiu Zeng

In Computer Networks and Communications

1. J. Langford
2. Yi Chang
3. Tomer Koren
4. Eric T. Bradlow
5. Eric M. Schwartz
6. Zhaohui Zheng
7. Oren Somekh
8. Peter S. Fader

In Management Science and Operations Research

1. Lihong Li
2. Nathan Kallus
3. J. Langford
4. Chunqiu Zeng
5. Liang Tang
6. Thorsten Joachims
7. Lei Li
8. Tao Li

Citations in the corpus

• A contextual-bandit approach to personalized news article recommendation (2010, neutral): “free algorithms [2] in real bucket tests.”
• Unbiased offline evaluation of contextual-bandit-based news article recommendation algorithms (2010, methodology): “Such Bayesian approaches may have competitive performance with appropriate prior distributions, but are often computationally pro-hibitive without coupling with approximation [2].”
• Online Context-Aware Recommendation with Time Varying Multi-Armed Bandit (2016, background): “The multi-armed bandit problem is identified in diverse applications, such as online advertising [20, 14], web content optimization [21, 1], and robotic routing [4].”
• Almost Optimal Exploration in Multi-Armed Bandits (2013, background): “These include online ad selection (Chakrabarti et al., 2008), web content optimization (Agarwal et al., 2009), usercontent matching (Pandey et al., 2007), learning search-ranking from usage data (Radlinski et al., 2008) and more.”
• Refining Recency Search Results with User Click Feedback (2011, methodology): “While models in earlier works ( e.g. , [1]) did not generalize, there have been efforts on generalization more recently, such as the LinUCB algorithm [21] that uses a similar linear model as ours, and the warm-start solution by [2].”
• Exploring or Exploiting? Social and Ethical Implications of Autonomous Experimentation in AI (2016, unknown): “Here, for example, the environment might be user attributes, the actions might be be various ad placements, and the reward might be increasing revenue [5, 17].”
• Customer Acquisition via Display Advertising Using Multi-Armed Bandit Experiments (2017, methodology): “This challenge relates to other work at the intersection of online advertising or online content optimization and MAB problems (Agarwal et al. 2008; Scott 2010; Urban et al. 2013).”
• Personalized Recommendation via Parameter-Free Contextual Bandits (2015, background): “A wide range of application-oriented problems have been modeled as the context-free multi-armed bandit problem, such as online advertising [25, 30], web search and content optimization [2, 27], network optimization [11], game playing [12], routing [6], etc.”
• Learning to model relatedness for news recommendation (2011, background): “In news domain, recommendation services often rely on user activities such as click patterns to find the most popular news [2], user-news ratings for collaborative news recommendation [25, 17, 10], or user interests for content-based and adaptive news access [4].”