25th Annual Conference on Learning Theory Analysis of Thompson Sampling for the Multi-armed Bandit Problem
Citations in the corpus
Neural Processes (2018, methodology): “It is commonly applied to black box optimisation and multi-armed bandit problems (e.g. Agrawal & Goyal, 2012; Shahriari et al., 2016).”
Learning to Optimize via Posterior Sampling (2013, background): “The Bayesian regret bounds we will establish are stronger than those implied by the results of Agrawal and Goyal [6].”
Efficient Exploration Through Bayesian Deep Q-Networks (2018, background): “In multi-arm bandit, there are compelling empirical pieces of evidence that Thompson Sampling sometimes provides better results than optimism-under-uncertainty approaches (Chapelle & Li, 2011), while also the performance guarantees are preserved (Agrawal & Goyal, 2012; Russo & Van Roy, 2014a).”
A Tutorial on Thompson Sampling (2017, positive): “A series of papers provided proofs confirming this finding [37, 38, 39].”
Deep Exploration via Randomized Value Functions (2017, background): “Further, the theory suggests that “well-designed” upper-confidence-bound-based approaches, which appropriately couple uncertainties across state-action pairs, but are often computationally intractable, are similarly near-optimal (statistically) and competitive with Thompson sampling in such…”
Thompson Sampling for Contextual Bandits with Linear Payoffs (2012, methodology): “For the the basic (i.e. without contexts) version of the stochastic MAB problem, some significant progress was made by Agrawal & Goyal (2012), Kaufmann et al. (2012) and, more recently, by Agrawal & Goyal (2013b), who provided optimal regret bounds on the expected regret.”
Double Thompson Sampling for Dueling Bandits (2016, methodology): “Inspired by the insights from Condorcet dueling bandits, we also discuss how to refine the regret of D-TS in general Copeland dueling bandits.”
Chameleon: scalable adaptation of video analytics (2018, methodology): “al [19] use Thompson sampling [11] to optimize the layout of a multi-component web page; they use greedy hill climbing to select the next layout, similar to Chameleon, but Chameleon exploits more independence structure and monotonicity.”
On Kernelized Multi-armed Bandits (2017, methodology): “Our second algorithm, GP-TS (Algorithm 2), inspired by the success of Thompson sampling for standard and parametric bandits (Agrawal and Goyal, 2012; Kaufmann et al., 2012; Gopalan et al., 2014; Agrawal and Goyal, 2013), uses the time-varying scale parameter v t = B + R 2( γ t − 1 + 1 + ln(2 /δ ))…”
From Bandits to Monte-Carlo Tree Search: The Optimistic Principle Applied to Optimization and Planning (2014, methodology): “…and an arm is selected as a function of the posterior (such as Thompson sampling [Thompson, 1933, 1935] which has been analyzed recently in [Agrawal and Goyal, 2012, Kauffmann et al., 2012, Agrawal and Goyal, 2013, Kaufmann et al., 2013], the Gittins indexes, see [Gittins., 1979, Gittins…”
Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Active Tasks (2018, methodology): “ commonly target the learning algorithm itself (e.g. PPO [67], Q-Learning [51], SAC [29], et cetera), how to efficiently explore the state space [23,58], or how to balance exploration and exploitation [3,5]. These can be seen as users of our method as we essentially update the input state from pixels”
Federated Recommendation System via Differential Privacy (2020, methodology): “Another mainstream approach is the sampling-based approach [10] that instead of computing a deterministic index, it uses a sample generated by a Bayesian estimator.”