Browse wiki

From MurrayWiki
Jump to: navigation, search
A group-theoretic approach to formalizing bootstrapping problems
Abstract The bootstrapping problem consists in desi
The bootstrapping problem consists in designing agents that laern a model of themsleves and the world, and utilize it to achieve useful tasks. It is different from other learning problems as the agent starts with uninterpreted observaions and commands, and with minimal prior information about the world. In this paper, we give a mathematical formalizatoin of this aspect of the problem. We argue that the vague constraint of having “no prior information” can be recast as a precise algebraic condition on the agent: that its behavior is invariant to particular classes of nuisances on the world, which we show can be well represented by actions of groups (diffeomorphisms, permutatations, linear transformations) on observations and commands. We then introduce the class of bilinear gradient dynamics sensors (DGDS) as a candidate for learning generic rootic sensorimotor cascades. We show how framing the problem as rejections of group nuisances allows a compact and modular analysis of typical preprocessing stages, such as learning the toplogy of sensors. We demonstrate learning and using such models on real-world range-finder and camera data from publicly available datasets.
era data from publicly available datasets.  +
Authors Andrea Censi and Richard M. Murray  +
ID 2011a  +
Source 2011 International Conference on Intelligent Robots and Systems (IROS)  +
Tag cm11-iros  +
Title A group-theoretic approach to formalizing bootstrapping problems +
Type Conference Paper  +
Categories Papers
Modification date
This property is a special property in this wiki.
15 May 2016 06:16:09  +
URL
This property is a special property in this wiki.
http://www.cds.caltech.edu/~murray/preprints/cm11-iros.pdf  +
hide properties that link here 
A group-theoretic approach to formalizing bootstrapping problems + Title
 

 

Enter the name of the page to start browsing from.