Optimistic Bayesian Sampling in Contextual-Bandit Problems

By Benedict C. May, N. Korda, Anthony Lee, David S. Leslie. 2012

139 citations recorded by the source corpus.

Subfields: Reinforcement Learning, Probabilistic & Bayesian Methods

How researchers see this paper

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

In Reinforcement Learning

In Probabilistic & Bayesian Methods

Citations in the corpus

Taking the Human Out of the Loop: A Review of Bayesian Optimization (2016, neutral): “[30], [50], [51] and contextual bandits [2], [97], [112].”
Learning to Optimize via Posterior Sampling (2013, background): “[30]), 1 D ow nl oa de d fr om in fo rm s.”
Thompson Sampling for Contextual Bandits with Linear Payoffs (2012, background): “Granmo (2010) and May et al. (2011) provided weak guarantees, namely, a bound of o ( T ) on the expected regret in time T .”
Bayesian Multi-Scale Optimistic Optimization (2014, methodology): “If we were to implement Thompson sampling strategies (May et al., 2011; Kaufmann et al., 2012; Agrawal & Goyal, 2013) for Gaussian processes (GPs), we would also encounter the difﬁcult problem of having to ﬁnd the maximizer of a sample from the GP at each iteration, unless we were considering only…”
Customer Acquisition via Display Advertising Using Multi-Armed Bandit Experiments (2017, methodology): “We obtain the distribution p(µj|Dt), and we can carry through our subscript j and then follow the procedures from the TS literature (Chapelle and Li 2011; Granmo 2010; May et al. 2011; Scott 2010).”
Online Learning: A Comprehensive Survey (2018, background): “terior distributions, and is able to achieve the lower bound of Lai and Robbins (1985) for Bernoulli rewards. More other extensive and recent studies of Bayesian Bandits can be found in (Scott, 2010; May et al., 2012; Russo and Van Roy, 2016). 5.3 Adversarial Bandits In the previous setting of stoch”
An Information-Theoretic Analysis of Thompson Sampling (2014, background): “One of the ﬁrst theoretical guarantees for Thompson sampling was provided by May et al. [22], but they showed only that the algorithm converges asymptotically to optimality.”
Further Optimal Regret Bounds for Thompson Sampling (2012, methodology): “Then the expected total regret in time T is given by P Other performance measures include PAC-style guarantees; we do not consider those measures here.”