Spectral Learning of Mixture of Hidden Markov Models

2014-12-01 · NeurIPS 2014 12 · Cem Subakan, Johannes Traa, Paris Smaragdis

In this paper, we propose a learning approach for the Mixture of Hidden Markov Models (MHMM) based on the Method of Moments (MoM). Computational advantages of MoM make MHMM learning amenable for large data sets. It is not possible to directly learn an MHMM with existing learning approaches, mainly due to a permutation ambiguity in the estimation process. We show that it is possible to resolve this ambiguity using the spectral properties of a global transition matrix even in the presence of estimation noise. We demonstrate the validity of our approach on synthetic and real data.

📄 PDF Abstract BibTeX

Code (0)

등록된 구현이 없습니다.

Analysis of Brain States from Multi-Region LFP Time-Series

2014-12-01 · NeurIPS 2014 12 · Kyle R. Ulrich, David E. Carlson, Wenzhao Lian, Jana S. Borg 외

The local field potential (LFP) is a source of information about the broad patterns of brain activity, and the frequencies present in these time-series measurements are often highly correlated between regions. It is beli…

Gaussian Processes Time Series Time Series Analysis

GP Kernels for Cross-Spectrum Analysis

2015-12-01 · NeurIPS 2015 12 · Kyle R. Ulrich, David E. Carlson, Kafui Dzirasa, Lawrence Carin

Multi-output Gaussian processes provide a convenient framework for multi-task problems. An illustrative and motivating example of a multi-task problem is multi-region electrophysiological time-series data, where experim…

Gaussian Processes Time Series Time Series Analysis

Model-based clustering with Hidden Markov Model regression for time series with regime changes

2013-12-25 · Faicel Chamroukhi, Allou Samé, Patrice Aknin, Gérard Govaert

This paper introduces a novel model-based clustering approach for clustering time series which present changes in regime. It consists of a mixture of polynomial regressions governed by hidden Markov chains. The underlyin…

Clustering model parameter estimation regression+3

Transform-Invariant Non-Parametric Clustering of Covariance Matrices and its Application to Unsupervised Joint Segmentation and Action Discovery

2017-10-27 · Nadia Figueroa, Aude Billard

In this work, we tackle the problem of transform-invariant unsupervised learning in the space of Covariance matrices and applications thereof. We begin by introducing the Spectral Polytope Covariance Matrix (SPCM) Simila…

Clustering Time Series Time Series Analysis

Spectral Methods for Learning Multivariate Latent Tree Structure

2011-12-01 · NeurIPS 2011 12 · Animashree Anandkumar, Kamalika Chaudhuri, Daniel J. Hsu, Sham M. Kakade 외

This work considers the problem of learning the structure of multivariate linear tree models, which include a variety of directed tree graphical models with continuous, discrete, and mixed latent variables such as linear…