Stochastic variational inference for hidden Markov models
By N. Foti, Jason Xu, Dillon Laird, Emily B. Fox. 2014
99 citations recorded by the source corpus.
Variational inference algorithms have proven successful for Bayesian analysis in large data settings, with recent advances using stochastic variational inference (SVI). However, such methods have largely been studied in independent or exchangeable data settings. We develop an SVI algorithm to learn the parameters of hidden Markov models (HMMs) in a time-dependent data setting. The challenge in applying stochastic optimization in this setting arises from dependencies in the chain, which must be broken to consider minibatches of observations. We propose an algorithm that harnesses the memory decay of the chain to adaptively bound errors arising from edge effects. We demonstrate the effectiveness of our algorithm on synthetic experiments and a large genomics dataset where a batch algorithm is computationally infeasible.
Subfields: Artificial Intelligence
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
- David M. Blei
- Stephan Mandt
- Yoshua Bengio
- Sergey Levine
- Dustin Tran
- Max Welling
- Qiang Liu
- Quoc V. Le
Citations in the corpus
- Variational Inference: A Review for Statisticians (2016, background): “…et al., 2014), general time-series models (Roberts et al., 2004; Barber and Chiappa, 2006; Archambeau et al., 2007b,a; Johnson and Willsky, 2014; Foti et al., 2014), robust models (Tipping and Lawrence, 2005; Wang and Blei, 2015), and Gaussian process models (Titsias and Lawrence, 2010;…”
- Hidden Markov Nonlinear ICA: Unsupervised Learning from Nonstationary Time Series (2020, methodology): “In such situations we can adapt the subchain sampling approach of Foti et al. (2014).”
- Edward: A library for probabilistic modeling, inference, and criticism (2016, neutral): “For criticism , Edward provides methods from scoring rules (Winkler, 1996) and predictive”
- Advances in Variational Inference (2017, methodology): “[45], [76] derive SVI for popular time series models including HMMs, hidden semi-Markov models (HSMM), and hierarchical Dirichlet process-HMMs.”
- Structured VAEs: Composing Probabilistic Graphical Models and Variational Autoencoders (2016, background): “It is also possible to sample subsequences and compute controllably-biased stochastic gradients (Foti et al., 2014).”
- Deep Probabilistic Programming (2017, background): “The approach extends naturally to streaming data (Doucet et al., 2000; Broderick et al., 2013; McInerney et al., 2015), dynamic batch sizes, and data structures in which working on a subgraph does not immediately apply (Binder et al., 1997; Johnson & Willsky, 2014; Foti et al., 2014).”