BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Infinite Hidden Semi-Markov Models - Carl Ashworth\; Vince Velkey
DTSTART:20260129T090000Z
DTEND:20260129T110000Z
UID:TALK243775@talks.cam.ac.uk
CONTACT:124819
DESCRIPTION:In this journal club\, we will explore how infinite hidden sem
 i-markov models\, which combine Bayesian nonparametrics with hidden Markov
  model-like dynamics\, can be used for automatic segmentation of behaviour
 s during learning. We will begin by providing an overview of infinite hidd
 en Semi-Markov models [2]\, and contrasting them with the classical infini
 te hidden Markov model [1]. Then\, extending this framework to a behaviour
 al modelling context\, we will present a recent nature neuroscience paper 
 by Bruins et al. [3]\, which uses an infinite hidden semi-Markov model to 
 capture the learning dynamics of individual mice in a large dataset as the
 y performed a perceptual decision making task. By allowing for modest vari
 ations in the existing states\, as well as the introduction of new states 
 throughout the task\, they demonstrate three main phases of learning acros
 s the population of mice\, alongside large inter-individual differences. O
 verall\, we aim to provide a view into Bayesian nonparametrics as a valuab
 le tool for automatic state segmentation and for the characterisation of i
 ndividual learning trajectories in a neuroscientific context.\n\n[1] Beal\
 , M. J.\, Ghahramani\, Z.\, & Rasmussen\, C. E. (2001). The infinite hidde
 n Markov model. Advances in Neural Information Processing Systems\, 14.\n[
 2] Johnson\, M. J.\, & Willsky\, A. S. (2013). Bayesian nonparametric hidd
 en semi-Markov models. Journal of Machine Learning Research\, 14\, 673–7
 01.\n[3] Bruijns\, S. A.\, Bougrova\, K.\, Laranjeira\, I. C.\, et al. (20
 26). Infinite hidden Markov models can dissect the complexities of learnin
 g. Nature Neuroscience\, 29(1)\, 186–194.
LOCATION:CBL Seminar Room\, Engineering Department\, 4th floor Baker build
 ing
END:VEVENT
END:VCALENDAR