Model selection performance of sPHMM in ground truth datasets.

A: Number of states inferred by sPHMM vs. the true number of states in MMPP datasets with 10 neurons and different numbers of hidden states (72 datasets in total, 3 datasets for each m) and 100 random initial guesses for the parameter values. The number...

Full description

Saved in:

Bibliographic Details
Main Author:	Tianshu Li (1527088) (author)
Other Authors:	Giancarlo La Camera (94285) (author)
Published:	2025
Subjects:	Biotechnology Cancer Inorganic Chemistry Infectious Diseases Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified surrogate datasets built rapid state switching bayesian information criterion alternative strategy based sphmm also captures uncovered hidden states hidden markov models potentially related problem cortical data requires hmm &# 8217 enforce large self hidden states sphmm ). since models large prior &# 8216 xlink "> used together transition probabilities statistical properties second problem neural dynamics neural data model selection many reports low self higher likelihood ground truth first problem cortex tend cognitive processes careful handling
Tags:	Add Tag No Tags, Be the first to tag this record!

Model selection performance of sPHMM in ground truth datasets.

Similar Items