TU Darmstadt / ULB / TUbiblio

Non-Parametric Bayesian Inference for Change Point Detection in Neural Spike Trains

Alt, Bastian and Messer, Michael and Roeper, Jochen and Schneider, Gaby and Koeppl, Heinz (2018):
Non-Parametric Bayesian Inference for Change Point Detection in Neural Spike Trains.
Freiburg im Breisgau, Germany, In: 2018 IEEE Statistical Signal Processing Workshop (SSP) (SSP 2018), Freiburg im Breisgau, Germany, [Conference or Workshop Item]

Abstract

We present a model for point processes with gamma distributed increments. We assume a piecewise constant latent process controlling shape and scale of the distribution. For the discrete number of states of the latent process we use a non-parametric assumption by utilizing a Chinese restaurant process (CRP). For the inference of such inhomogeneous gamma processes with an unbounded number of states we do Bayesian inference using Markov Chain Monte Carlo. Finally, we apply the inference algorithm to simulated point processes and to empirical spike train recordings, which inherently possess non-stationary and non-Poissonian behavior.

Item Type: Conference or Workshop Item
Erschienen: 2018
Creators: Alt, Bastian and Messer, Michael and Roeper, Jochen and Schneider, Gaby and Koeppl, Heinz
Title: Non-Parametric Bayesian Inference for Change Point Detection in Neural Spike Trains
Language: English
Abstract:

We present a model for point processes with gamma distributed increments. We assume a piecewise constant latent process controlling shape and scale of the distribution. For the discrete number of states of the latent process we use a non-parametric assumption by utilizing a Chinese restaurant process (CRP). For the inference of such inhomogeneous gamma processes with an unbounded number of states we do Bayesian inference using Markov Chain Monte Carlo. Finally, we apply the inference algorithm to simulated point processes and to empirical spike train recordings, which inherently possess non-stationary and non-Poissonian behavior.

Place of Publication: Freiburg im Breisgau, Germany
Uncontrolled Keywords: Inhomogeneous Gamma Process; Bayesian Non-Parametrics; Neural Spike Trains; Change Points
Divisions: 18 Department of Electrical Engineering and Information Technology > Institute for Telecommunications > Bioinspired Communication Systems
DFG-Collaborative Research Centres (incl. Transregio) > Collaborative Research Centres > CRC 1053: MAKI – Multi-Mechanisms Adaptation for the Future Internet
DFG-Collaborative Research Centres (incl. Transregio) > Collaborative Research Centres > CRC 1053: MAKI – Multi-Mechanisms Adaptation for the Future Internet > B: Adaptation Mechanisms > Subproject B4: Planning
18 Department of Electrical Engineering and Information Technology > Institute for Telecommunications
DFG-Collaborative Research Centres (incl. Transregio) > Collaborative Research Centres
DFG-Collaborative Research Centres (incl. Transregio) > Collaborative Research Centres > CRC 1053: MAKI – Multi-Mechanisms Adaptation for the Future Internet > B: Adaptation Mechanisms
18 Department of Electrical Engineering and Information Technology
DFG-Collaborative Research Centres (incl. Transregio)
Event Title: 2018 IEEE Statistical Signal Processing Workshop (SSP) (SSP 2018)
Event Location: Freiburg im Breisgau, Germany
Date Deposited: 26 Apr 2018 22:35
Export:

Optionen (nur für Redakteure)

View Item View Item