Information Bottleneck Approach to Predictive Inference

Information Bottleneck Approach to Predictive Inference

This paper synthesizes a recent line of work on automated predictive model making inspired by Rate-Distortion theory, in particular by the Information Bottleneck method. Predictive inference is interpreted as a strategy for efficient communication. The relationship to thermodynamic efficiency is dis...

Full description

Bibliographic Details
Main Author: Susanne Still
Format: Article
Language:English
Published: MDPI AG 2014-02-01
Series:Entropy
Subjects:
predictive inference
information bottleneck method
dynamical systems
thermodynamic efficiency
far-from-equilibrium thermodynamics
computing engines
Online Access:http://www.mdpi.com/1099-4300/16/2/968
id doaj-4b0270313d604288af1aef9174261307
record_format Article
spelling doaj-4b0270313d604288af1aef91742613072020-11-24T20:40:36ZengMDPI AGEntropy1099-43002014-02-0116296898910.3390/e16020968e16020968Information Bottleneck Approach to Predictive InferenceSusanne Still0Information and Computer Sciences Department, University of Hawaii at Mānoa, Honolulu, HI 96822, USAThis paper synthesizes a recent line of work on automated predictive model making inspired by Rate-Distortion theory, in particular by the Information Bottleneck method. Predictive inference is interpreted as a strategy for efficient communication. The relationship to thermodynamic efficiency is discussed. The overall aim of this paper is to explain how this information theoretic approach provides an intuitive, overarching framework for predictive inference.http://www.mdpi.com/1099-4300/16/2/968predictive inferenceinformation bottleneck methoddynamical systemsthermodynamic efficiencyfar-from-equilibrium thermodynamicscomputing engines
collection DOAJ
language English
format Article
sources DOAJ
author Susanne Still
spellingShingle Susanne Still
Information Bottleneck Approach to Predictive Inference
Entropy
predictive inference
information bottleneck method
dynamical systems
thermodynamic efficiency
far-from-equilibrium thermodynamics
computing engines
author_facet Susanne Still
author_sort Susanne Still
title Information Bottleneck Approach to Predictive Inference
title_short Information Bottleneck Approach to Predictive Inference
title_full Information Bottleneck Approach to Predictive Inference
title_fullStr Information Bottleneck Approach to Predictive Inference
title_full_unstemmed Information Bottleneck Approach to Predictive Inference
title_sort information bottleneck approach to predictive inference
publisher MDPI AG
series Entropy
issn 1099-4300
publishDate 2014-02-01
description This paper synthesizes a recent line of work on automated predictive model making inspired by Rate-Distortion theory, in particular by the Information Bottleneck method. Predictive inference is interpreted as a strategy for efficient communication. The relationship to thermodynamic efficiency is discussed. The overall aim of this paper is to explain how this information theoretic approach provides an intuitive, overarching framework for predictive inference.
topic predictive inference
information bottleneck method
dynamical systems
thermodynamic efficiency
far-from-equilibrium thermodynamics
computing engines
url http://www.mdpi.com/1099-4300/16/2/968
work_keys_str_mv AT susannestill informationbottleneckapproachtopredictiveinference
_version_ 1716826330524286976

Similar Items