On‐line inference for multiple changepoint problems

By Paul Fearnhead, Z. Liu. 2007

367 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Genetics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/688581bc42b30458f8ac951ea31d6fe5e4df164a

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

In Genetics

In Statistics and Probability

Citations in the corpus

EXACT ASYMPTOTIC DISTRIBUTION OF CHANGE-POINT MLE FOR CHANGE IN THE MEAN OF GAUSSIAN SEQUENCES (2010, background): “Even though there are recent advances in addressing multiple changes in scientific phenomena [see Fearnhead (2006), Fearnhead and Liu (2007), Girón, Moreno and Casella (2007) and Seidou and Ouarda (2007)], the classical change-point literature is most well developed in the case of a single unknown…”
An Algorithmic Perspective on Imitation Learning (2018, methodology): “As seen from these previous studies, advances in trajectory segmentation in imitation learning [Kulić et al., 2008, Konidaris et al., 2011, Niekum et al., 2014] are closely related to the methodological advances [Fearnhead and Liu, 2007, Fox et al., 2009] in the machine learning community.”
Human motion trajectory prediction: a survey (2019, methodology): “Galceran et al. (2015) introduce a multi-policy decision-making systems to generate robot motions based on predicted movements of other agents in the scene, estimated with a changepoint-based technique (Fearnhead and Liu 2007).”
Explicit-Duration Markov Switching Models (2014, background): “More recent Bayesian approaches define a distribution on the number and positions of changepoints through a segmentduration distribution [Fearnhead, 2006, Fearnhead and Liu, 2007, Adams and MacKay, 2007, Fearnhead and Vasileiou, 2009, Eckley et al., 2011].”
Continuous Meta-Learning without Tasks (2019, methodology): “The Bayesian online changepoint framework of Adams & MacKay (2007) (which we leverage in this paper) and the similar, simultaneous work of Fearnhead & Liu (2007) have generated a substantial body of follow-on work since their publication.”
Robot learning from demonstration by constructing skill trees (2012, methodology): “Fearnhead and Liu (2007) introduced online algorithms for both Bayesian and maximum a posteriori (MAP) changepoint detection.”
Adaptive methods for sequential importance sampling with application to state space models (2008, background): “Fearnhead and Liu (2007, Section 3.2) adjust the size of the particle cloud in order to control the error introduced by the resampling step.”
Deep Reinforcement Learning amidst Lifelong Non-Stationarity (2020, background): “In particular, this notion of infrequent, discrete shifts underlies the changepoint detection literature [1, 13].”
Biips: Software for Bayesian Inference with Interacting Particle Systems (2014, methodology): “However, SMC are far from been restricted to this class of models and have been used more broadly, either alone or as part of a MCMC algorithm (Fearnhead 2004; Fearnhead and Liu 2007; Caron et al. 2008; Andrieu et al. 2010; Caron et al. 2012; Naesseth et al. 2014).”
Sequential Bayesian Prediction in the Presence of Changepoints and Faults (2010, neutral): “Sequential and dependent changepoint detection has been performed [11] only for a limited set of changepoint models.”
Constructing Skill Trees for Reinforcement Learning Agents from Demonstration Trajectories (2010, methodology): “Following Fearnhead and Liu [8], we assume conjugate priors: the Gaussian noise prior has mean zero, and variance with inverse gamma prior with parameters v2 and u 2 .”
Deep Reinforcement Learning amidst Continual Structured Non-Stationarity (2021, methodology): “For highly infrequent shifts however, we may need to leverage alternative tools; in particular, this notion of infrequent, discrete shifts underlies the changepoint detection literature (Adams & MacKay, 2007; Fearnhead & Liu, 2007).”
Quantifying the uncertainty in change points (2012, methodology): “This involves sampling from the joint posterior distribution of the model parameters and the underlying state sequence (see e.g. Chib, 1998; Fearnhead, 2006; Fearnhead and Liu, 2007).”