Thesis Supervisor Accepted by.......................................................................

2001

Subfields: Social Sciences & Humanities

Citations in the corpus

• Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness (2020, background): “, expectation propagation (EP)) can be applied (Minka, 2001).”
• A General Framework for Uncertainty Estimation in Deep Learning (2019, background): “[23] showed that the solution to (7) requires matching the moments of the two distributions.”
• Kernel-Based Just-In-Time Learning for Passing Expectation Propagation Messages (2015, background): “We address the particular setting of expectation propagation (Minka, 2001), a message passing algorithm wherein messages are confined to being members of a particular parametric family.”
• GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learning (2021, methodology): “Some classic non-augmentation based methods include the Laplace approximation (Williams & Barber, 1998), expectation-propagation (Minka, 2001), and the least square approach (Rifkin & Klautau, 2004).”
• Deep Bayesian Bandits Showdown: An Empirical Comparison of Bayesian Deep Networks for Thompson Sampling (2018, background): “The family of expectation-propagation algorithms (Opper & Winther, 2000; Minka, 2001b;a) is based on the message passing framework (Pearl, 1986).”
• Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks (2015, methodology): “The next section provides a detailed description of PBP, presenting it as an assumed density filtering (ADF) method (Opper & Winther, 1998) that uses some of the improvements on ADF given by expectation propagation (Minka, 2001).”
• Expectation Propagation for approximate Bayesian inference (2001, background): “Regular EP did not converge, but a restricted version did (Minka, 2001).”
• Streaming Variational Bayes (2013, methodology): “This approach is similar in spirit to assumed density filtering or expectation propagation [7–9], but each step of those methods involves a moment-matching step that can be computationally costly for models such as topic models.”
• Fixed-Form Variational Posterior Approximation through Stochastic Linear Regression (2012, background): “For Gaussian q ( x ) and twice differentiable log p ( x, y ), Minka (2001) and Opper and Archambeau (2009) show that and where ∇ x ∇ x log p ( x, y ) denotes the Hessian matrix of log p ( x, y ) in x .”
• Black-box $\alpha$-divergence Minimization (2015, methodology): “Then according to (Minka, 2001a), the EP energy function is In practice EP finds a stationary solution to the constrained optimization problem where the constraint in (15) guarantees that the { λ \ n } are valid cavity parameters that are consistent with the approximate posterior.”
• Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Control (2017, methodology): “In this paper, however, we exploit the fact that the moment-matching approximation (8) is a deterministic operator, similar to the projection used in EP [30, 23].”
• Predictive Entropy Search for Efficient Global Optimization of Black-box Functions (2014, background): “In what follows we briefly show how to construct these factors; more detail is given in Appendix B.”
• Robot trajectory optimization using approximate inference (2009, methodology): “Expectation Propagation (Minka, 2001b) is a method to iteratively improve the messages (and thereby the belief) when they are bound to be approximate.”
• Learning to Pass Expectation Propagation Messages (2013, methodology): “Given incoming messages {miψ(xi)}i∈Scope(ψ), the simplest approach to computing mψi(xi) is to use importance sampling.”