A world full of surprises: Bayesian theory of surprise to quantify degrees of uncertainty

Nelly Bencomo, Amel Belaggoun

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the specific area of software engineering (SE) for self-adaptive systems (SASs) there is a growing research awareness about the synergy between SE and artificial intelligence (AI). However, just few significant results have been published so far. In this paper, we propose a novel and formal Bayesian definition of surprise as the basis for quantitative analysis to measure degrees of uncertainty and deviations of self-adaptive systems from normal behavior. A surprise measures how observed data affects the models or assumptions of the world during runtime. The key idea is that a "surprising" event can be defined as one that causes a large divergence between the belief distributions prior to and posterior to the event occurring. In such a case the system may decide either to adapt accordingly or to flag that an abnormal situation is happening. In this paper, we discuss possible applications of Bayesian theory of surprise for the case of self-adaptive systems using Bayesian dynamic decision networks.

Original languageEnglish
Title of host publicationCompanion : proceedings of the 36th International Conference on Software Engineering
Place of PublicationNew York, NY (US)
PublisherACM
Pages460-463
Number of pages4
ISBN (Print)978-1-4503-2768-8
DOIs
Publication statusPublished - 31 Dec 2014
Event36th International Conference on Software Engineering - Hyderabad, India
Duration: 31 May 20147 Jun 2014

Conference

Conference36th International Conference on Software Engineering
Abbreviated titleICSE 2014
CountryIndia
CityHyderabad
Period31/05/147/06/14

Fingerprint

Adaptive systems
Software engineering
Artificial intelligence
Chemical analysis
Uncertainty

Keywords

  • Bayesian networks
  • Bayesian surprise
  • self-adaptation
  • uncertainty

Cite this

Bencomo, N., & Belaggoun, A. (2014). A world full of surprises: Bayesian theory of surprise to quantify degrees of uncertainty. In Companion : proceedings of the 36th International Conference on Software Engineering (pp. 460-463). New York, NY (US): ACM. https://doi.org/10.1145/2591062.2591118
Bencomo, Nelly ; Belaggoun, Amel. / A world full of surprises : Bayesian theory of surprise to quantify degrees of uncertainty. Companion : proceedings of the 36th International Conference on Software Engineering. New York, NY (US) : ACM, 2014. pp. 460-463
@inproceedings{1c2f69cd99bd4f87afef3f7ff21566dc,
title = "A world full of surprises: Bayesian theory of surprise to quantify degrees of uncertainty",
abstract = "In the specific area of software engineering (SE) for self-adaptive systems (SASs) there is a growing research awareness about the synergy between SE and artificial intelligence (AI). However, just few significant results have been published so far. In this paper, we propose a novel and formal Bayesian definition of surprise as the basis for quantitative analysis to measure degrees of uncertainty and deviations of self-adaptive systems from normal behavior. A surprise measures how observed data affects the models or assumptions of the world during runtime. The key idea is that a {"}surprising{"} event can be defined as one that causes a large divergence between the belief distributions prior to and posterior to the event occurring. In such a case the system may decide either to adapt accordingly or to flag that an abnormal situation is happening. In this paper, we discuss possible applications of Bayesian theory of surprise for the case of self-adaptive systems using Bayesian dynamic decision networks.",
keywords = "Bayesian networks, Bayesian surprise, self-adaptation, uncertainty",
author = "Nelly Bencomo and Amel Belaggoun",
year = "2014",
month = "12",
day = "31",
doi = "10.1145/2591062.2591118",
language = "English",
isbn = "978-1-4503-2768-8",
pages = "460--463",
booktitle = "Companion : proceedings of the 36th International Conference on Software Engineering",
publisher = "ACM",
address = "United States",

}

Bencomo, N & Belaggoun, A 2014, A world full of surprises: Bayesian theory of surprise to quantify degrees of uncertainty. in Companion : proceedings of the 36th International Conference on Software Engineering. ACM, New York, NY (US), pp. 460-463, 36th International Conference on Software Engineering, Hyderabad, India, 31/05/14. https://doi.org/10.1145/2591062.2591118

A world full of surprises : Bayesian theory of surprise to quantify degrees of uncertainty. / Bencomo, Nelly; Belaggoun, Amel.

Companion : proceedings of the 36th International Conference on Software Engineering. New York, NY (US) : ACM, 2014. p. 460-463.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A world full of surprises

T2 - Bayesian theory of surprise to quantify degrees of uncertainty

AU - Bencomo, Nelly

AU - Belaggoun, Amel

PY - 2014/12/31

Y1 - 2014/12/31

N2 - In the specific area of software engineering (SE) for self-adaptive systems (SASs) there is a growing research awareness about the synergy between SE and artificial intelligence (AI). However, just few significant results have been published so far. In this paper, we propose a novel and formal Bayesian definition of surprise as the basis for quantitative analysis to measure degrees of uncertainty and deviations of self-adaptive systems from normal behavior. A surprise measures how observed data affects the models or assumptions of the world during runtime. The key idea is that a "surprising" event can be defined as one that causes a large divergence between the belief distributions prior to and posterior to the event occurring. In such a case the system may decide either to adapt accordingly or to flag that an abnormal situation is happening. In this paper, we discuss possible applications of Bayesian theory of surprise for the case of self-adaptive systems using Bayesian dynamic decision networks.

AB - In the specific area of software engineering (SE) for self-adaptive systems (SASs) there is a growing research awareness about the synergy between SE and artificial intelligence (AI). However, just few significant results have been published so far. In this paper, we propose a novel and formal Bayesian definition of surprise as the basis for quantitative analysis to measure degrees of uncertainty and deviations of self-adaptive systems from normal behavior. A surprise measures how observed data affects the models or assumptions of the world during runtime. The key idea is that a "surprising" event can be defined as one that causes a large divergence between the belief distributions prior to and posterior to the event occurring. In such a case the system may decide either to adapt accordingly or to flag that an abnormal situation is happening. In this paper, we discuss possible applications of Bayesian theory of surprise for the case of self-adaptive systems using Bayesian dynamic decision networks.

KW - Bayesian networks

KW - Bayesian surprise

KW - self-adaptation

KW - uncertainty

UR - http://www.scopus.com/inward/record.url?scp=84903647929&partnerID=8YFLogxK

U2 - 10.1145/2591062.2591118

DO - 10.1145/2591062.2591118

M3 - Conference contribution

AN - SCOPUS:84903647929

SN - 978-1-4503-2768-8

SP - 460

EP - 463

BT - Companion : proceedings of the 36th International Conference on Software Engineering

PB - ACM

CY - New York, NY (US)

ER -

Bencomo N, Belaggoun A. A world full of surprises: Bayesian theory of surprise to quantify degrees of uncertainty. In Companion : proceedings of the 36th International Conference on Software Engineering. New York, NY (US): ACM. 2014. p. 460-463 https://doi.org/10.1145/2591062.2591118