BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
BEGIN:VEVENT
UID:533@lincs.fr
DTSTART;TZID=Europe/Paris:20200311T140000
DTEND;TZID=Europe/Paris:20200311T150000
DTSTAMP:20200312T095101Z
URL:https://www.lincs.fr/events/comparing-the-modelling-powers-of-rnn-and-
 hmm/
SUMMARY:Comparing the modelling powers of RNN and HMM
DESCRIPTION:Recurrent Neural Networks (RNN) and Hidden&nbsp\;Markov Models
 (HMM) are popular models for processing sequential data and have found many
 applications such as speech recognition\, time series prediction or machine
 translation. Although both models have been extended in several ways (eg.
 Long Short Term Memory and Gated Recurrent Unit architectures\, Variational
 RNN\, partially observed Markov models)\, their theoretical understanding
 remains partially open. In this context\, our approach consists in
 classifying both models from an information geometry point of view. More
 precisely\, both\nmodels can be used for modeling the distribution of a
 sequence of random observations from a set of latent variables\; however\,
 in RNN the latent variable is deterministically deduced from the current
 observation and the previous latent variable\, while in HMM the set of
 (random) latent variables is a Markov chain. In this paper\, we first embed
 these two generative models into a generative unified model (GUM). We next
 consider the subclass of GUM models which yield a stationary Gaussian
 observations probability distribution function (pdf). Such pdf are
 characterized by their covariance sequence\; we show that the GUM model can
 produce any stationary Gaussian distribution with geometrical covariance
 structure. We finally discuss about the modeling power of the HMM and RNN
 submodels\, via their associated observations pdf: some observations pdf
 can be modeled by a RNN\, but not by an HMM\, and vice versa\; some can be
 produced by both structures\, up to a reparameterization.
CATEGORIES:Seminars,Youtube
LOCATION:Paris-Rennes Room (EIT Digital)\, 23 avenue d'Italie\, 75013
 Paris\, France
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=23 avenue d'Italie\, 75013
 Paris\, France;X-APPLE-RADIUS=100;X-TITLE=Paris-Rennes Room (EIT
 Digital):geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20191027T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR