Lab in INRIA Lille, France.
The topic is time series prediction, and non-parametric statistical
analysis of time series.
Position:
3 years, fully funded position.
The position will start in September 2010. The deadline for applications
will be around mid March,
but interested candidates are encouraged to contact me earlier.
Environment:
The PhD student will work with Daniil Ryabko http://daniil.ryabko.net ,
and with other members of the SequeL group
http://sequel.futurs.inria.fr/ at INRIA Lille.
SequeL is one of the most dynamic labs at INRIA, with over 20
researchers (including PhD students)
working on both fundamental and practical aspects of sequential learning
problems:
from statistical learning, through reinforcement learning, to computer
poker and Go players.
INRIA is France's leading institution in Computer Science, with over
2800 scientists
employed, of which around 250 in Lille.
Lille is the capital of the north of France, a metropolis with 1 million
inhabitants,
with excellent train connection to Brussels (40 min), Paris (1h) and
London (1.5h by train).
Subject:
The basic problem setup is as follows. There is an unknown stochastic
source of data, generating observations in a sequential fashion. The
data can be anything from stock market observations, to DNA sequences,
to behavioural sequences. There are several learning and inference
problems connected with it, of which the two most basic ones are:
predicting the probabilities of the next observations, and testing
hypotheses about the source (such as independence, homogeneity, etc.)
To solve these problems, one has to consider models of the data.
Different types of data require different models.
The goal is to describe those probabilistic models under which
successful learning is possible, for the inference problems
considered: sequential prediction, and hypothesis testing. This would
eventually lead to an automated modelling algorithms for sequential
learning. The primary goal, however, is to establish a theoretical
understanding of what is possible to learn, in the sequential problems
of inference considered, under which assumptions. More precisely, a data
source is a probability distribution on the set of all possible
sequences of observations, and a model is a set of such probability
distributions. We are interested in identifying the properties of models
which ensure the existence of efficient algorithms that are successful
(e.g. as predictors) given the model.
Requirements:
The successful candidate will have a MSc or equivalent degree in
mathematics or computer science,
with strong background in probability and statistics.
Programming skills will be considered a plus.
The working language in the lab is English.
For further information please email daniil.ryabko at inria.fr ,
with subject -SeqPHD-, joining a CV and a description of interests.
See also http://www.inria.fr/travailler/opportunites/doc.en.html
**********************************************************
*
* Contributions to be spread via DMANET are submitted to
*
* DMANET@zpr.uni-koeln.de
*
* Replies to a message carried on DMANET should NOT be
* addressed to DMANET but to the original sender. The
* original sender, however, is invited to prepare an
* update of the replies received and to communicate it
* via DMANET.
*
* DISCRETE MATHEMATICS AND ALGORITHMS NETWORK (DMANET)
* http://www.zaik.uni-koeln.de/AFS/publications/dmanet/
*
**********************************************************
Posts
