[Comp-neuro] Probabilistic Reasoning and Decision Making in Sensory-Motor Systems - NEW BOOK

Jean-Marc Bollon Jean-Marc.Bollon at inrialpes.fr
Mon Sep 1 17:30:10 CEST 2008

Probabilistic Reasoning and Decision Making in Sensory-Motor Systems


Both living organisms and robotic systems must face the same central  
difficulty: How to survive being ignorant? How to use an incomplete  
and uncertain model of their environment to perceive, infer, decide,  
learn and act efficiently?

Indeed, any model of a real phenomenon is incomplete: there are always  
some hidden variables, not taken into account in the model, that  
influence the phenomenon. The effect of these hidden variables is that  
the model and the phenomenon never behave exactly alike. Uncertainty  
is the direct and unavoidable consequence of incompleteness. A model  
may not foresee exactly the future observations of a phenomenon as  
these observations are biased by the hidden variables. It may neither  
predict exactly the consequences of its decisions.

Probability theory, considered as an alternative to logic to model  
rational reasoning, is the perfect mathematical framework to face this  
difficult challenge. Learning is used in a first step to transform  
incompleteness into uncertainty, inference is then used to reason and  
take decisions based on the probability distributions constructed by  
learning. This so-called subjectivist approach to probability allows  
uncertain reasoning as complex and formal as the ones made using logic  
with exact knowledge.

This book presents twelve different implementations of this approach  
to very different sensory-motor systems either by programming robots  
or by modeling living systems.

Each of these works summarizes a PhD dissertation defended in  
different European universities.

All these works use Bayesian Programming: a mathematical formalism,  
which defines in simple mathematical terms the way probability, can be  
used as an alternative to logic. Bayesian Programming also proposes a  
programming and modeling methodology as, to respect the mathematical  
formalism, the programmer should follow always the same steps to build  
his model. Finally, Bayesian Programming is a common language to  
understand and compare the different models. This language is used all  
along this book by all the authors and insures the global coherence of  
these twelve very different examples.

More information : http://emotion.inrialpes.fr/BP/spip.php?article18

How to buy :



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080901/50097767/attachment.html

More information about the Comp-neuro mailing list