Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach

Andreas Holzinger, Markus Plass, Katharina Holzinger, Gloria Cerasela Crisan, Camelia M Pintea, Vasile Palade

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

Abstract

Most Machine Learning (ML) researchers focus on automatic Machine Learning (aML) where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from the availability of "big data". However, sometimes, for example in health informatics, we are confronted not a small number of data sets or rare events, and with complex problems where aML-approaches fail or deliver unsatisfactory results. Here, interactive Machine Learning (iML) may be of help and the "human-in-the-loop" approach may be beneficial in solving computationally hard problems, where human expertise can help to reduce an exponential search space through heuristics.
In this paper, experiments are discussed which help to evaluate the effectiveness of the iML-"human-in-the-loop" approach, particularly in opening the "black box", thereby enabling a human to directly and indirectly manipulating and interacting with an algorithm. For this purpose, we selected the Ant Colony Optimization (ACO) framework, and use it on the Traveling Salesman Problem (TSP) which is of high importance in solving many practical problems in health informatics, e.g. in the study of proteins.
LanguageEnglish
Title of host publicationLecture Notes in Computer Science, LNCS 9817
Subtitle of host publicationIFIP International Cross Domain Conference and Workshop (CD-ARES 2016)
PublisherSpringer
Pages81-95
DOIs
StatusPublished - 17 Aug 2016

Fingerprint

Traveling salesman problem
Learning systems
Health
Ant colony optimization
Recommender systems
Speech recognition
Availability
Proteins
Experiments

Keywords

  • Machine Learning
  • interactive Machine Learning
  • Human-in-the-loop
  • Ant-Colony Optimization
  • Traveling Salesman Problem

ASJC Scopus subject areas

  • Artificial Intelligence

Fields of Expertise

  • Information, Communication & Computing

Treatment code (Nähere Zuordnung)

  • Basic - Fundamental (Grundlagenforschung)

Cite this

Holzinger, A., Plass, M., Holzinger, K., Crisan, G. C., Pintea, C. M., & Palade, V. (2016). Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach. In Lecture Notes in Computer Science, LNCS 9817: IFIP International Cross Domain Conference and Workshop (CD-ARES 2016) (pp. 81-95). Springer. DOI: 10.1007/978-3-319-45507-5_6

Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach. / Holzinger, Andreas; Plass, Markus; Holzinger, Katharina; Crisan, Gloria Cerasela; Pintea, Camelia M; Palade, Vasile.

Lecture Notes in Computer Science, LNCS 9817: IFIP International Cross Domain Conference and Workshop (CD-ARES 2016). Springer, 2016. p. 81-95.

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

Holzinger, A, Plass, M, Holzinger, K, Crisan, GC, Pintea, CM & Palade, V 2016, Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach. in Lecture Notes in Computer Science, LNCS 9817: IFIP International Cross Domain Conference and Workshop (CD-ARES 2016). Springer, pp. 81-95. DOI: 10.1007/978-3-319-45507-5_6
Holzinger A, Plass M, Holzinger K, Crisan GC, Pintea CM, Palade V. Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach. In Lecture Notes in Computer Science, LNCS 9817: IFIP International Cross Domain Conference and Workshop (CD-ARES 2016). Springer. 2016. p. 81-95. Available from, DOI: 10.1007/978-3-319-45507-5_6
Holzinger, Andreas ; Plass, Markus ; Holzinger, Katharina ; Crisan, Gloria Cerasela ; Pintea, Camelia M ; Palade, Vasile. / Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach. Lecture Notes in Computer Science, LNCS 9817: IFIP International Cross Domain Conference and Workshop (CD-ARES 2016). Springer, 2016. pp. 81-95
@inproceedings{f60845c5c894473eb8a257e637d23183,
title = "Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach",
abstract = "Most Machine Learning (ML) researchers focus on automatic Machine Learning (aML) where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from the availability of {"}big data{"}. However, sometimes, for example in health informatics, we are confronted not a small number of data sets or rare events, and with complex problems where aML-approaches fail or deliver unsatisfactory results. Here, interactive Machine Learning (iML) may be of help and the {"}human-in-the-loop{"} approach may be beneficial in solving computationally hard problems, where human expertise can help to reduce an exponential search space through heuristics. In this paper, experiments are discussed which help to evaluate the effectiveness of the iML-{"}human-in-the-loop{"} approach, particularly in opening the {"}black box{"}, thereby enabling a human to directly and indirectly manipulating and interacting with an algorithm. For this purpose, we selected the Ant Colony Optimization (ACO) framework, and use it on the Traveling Salesman Problem (TSP) which is of high importance in solving many practical problems in health informatics, e.g. in the study of proteins.",
keywords = "Machine Learning, interactive Machine Learning, Human-in-the-loop, Ant-Colony Optimization, Traveling Salesman Problem",
author = "Andreas Holzinger and Markus Plass and Katharina Holzinger and Crisan, {Gloria Cerasela} and Pintea, {Camelia M} and Vasile Palade",
year = "2016",
month = "8",
day = "17",
doi = "10.1007/978-3-319-45507-5_6",
language = "English",
pages = "81--95",
booktitle = "Lecture Notes in Computer Science, LNCS 9817",
publisher = "Springer",

}

TY - GEN

T1 - Towards interactive Machine Learning (iML): Applying Ant Colony Algorithms to solve the Traveling Salesman Problem with the Human-in-the-Loop approach

AU - Holzinger,Andreas

AU - Plass,Markus

AU - Holzinger,Katharina

AU - Crisan,Gloria Cerasela

AU - Pintea,Camelia M

AU - Palade,Vasile

PY - 2016/8/17

Y1 - 2016/8/17

N2 - Most Machine Learning (ML) researchers focus on automatic Machine Learning (aML) where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from the availability of "big data". However, sometimes, for example in health informatics, we are confronted not a small number of data sets or rare events, and with complex problems where aML-approaches fail or deliver unsatisfactory results. Here, interactive Machine Learning (iML) may be of help and the "human-in-the-loop" approach may be beneficial in solving computationally hard problems, where human expertise can help to reduce an exponential search space through heuristics. In this paper, experiments are discussed which help to evaluate the effectiveness of the iML-"human-in-the-loop" approach, particularly in opening the "black box", thereby enabling a human to directly and indirectly manipulating and interacting with an algorithm. For this purpose, we selected the Ant Colony Optimization (ACO) framework, and use it on the Traveling Salesman Problem (TSP) which is of high importance in solving many practical problems in health informatics, e.g. in the study of proteins.

AB - Most Machine Learning (ML) researchers focus on automatic Machine Learning (aML) where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from the availability of "big data". However, sometimes, for example in health informatics, we are confronted not a small number of data sets or rare events, and with complex problems where aML-approaches fail or deliver unsatisfactory results. Here, interactive Machine Learning (iML) may be of help and the "human-in-the-loop" approach may be beneficial in solving computationally hard problems, where human expertise can help to reduce an exponential search space through heuristics. In this paper, experiments are discussed which help to evaluate the effectiveness of the iML-"human-in-the-loop" approach, particularly in opening the "black box", thereby enabling a human to directly and indirectly manipulating and interacting with an algorithm. For this purpose, we selected the Ant Colony Optimization (ACO) framework, and use it on the Traveling Salesman Problem (TSP) which is of high importance in solving many practical problems in health informatics, e.g. in the study of proteins.

KW - Machine Learning

KW - interactive Machine Learning

KW - Human-in-the-loop

KW - Ant-Colony Optimization

KW - Traveling Salesman Problem

U2 - 10.1007/978-3-319-45507-5_6

DO - 10.1007/978-3-319-45507-5_6

M3 - Conference contribution

SP - 81

EP - 95

BT - Lecture Notes in Computer Science, LNCS 9817

PB - Springer

ER -