Development of user-adaptive value system of learning function using interactive EC

Yuki Suga, Shigeki Sugano, Yoshinori Ikuma, Tetsuya Ogata

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

Abstract

Our goal is to create a user-adaptive communication-robot. We are developing a system for evaluating human-robot interactions. Although such evaluation is indispensable for learning algorithms, users' preferences are too difficult to model because they are subjective. In this study, we used the interactive evolutionary computation (IEC) to configure the value system of a learning communicationrobot. The IEC is a genetic algorithm whose fitness function is performed by the user. In our experiment, we encoded the values of sensors (reward or punishment) into genes, and subjects interacted with the learning robot. Through the interaction, the subjects evaluated the robot by touching its sensors, and the robot learned appropriate combinations between input and output. Afterward, the subjects gave their scores to the experimenter, and the scores were regarded as the fitness values of the corresponding genes. These sequences were continued until the 4 generation, and then the subjects compared three of their best genes and two of the experimenter's. We found that the user-adaptive value system is suitable for the communication-robot.

Original languageEnglish
Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
Volume17
Edition1 PART 1
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event17th World Congress, International Federation of Automatic Control, IFAC - Seoul
Duration: 2008 Jul 62008 Jul 11

Other

Other17th World Congress, International Federation of Automatic Control, IFAC
CitySeoul
Period08/7/608/7/11

Fingerprint

Robots
Genes
Evolutionary algorithms
Robot learning
Human robot interaction
Communication
Sensors
Learning algorithms
Genetic algorithms
Experiments

Keywords

  • Intelligent robotics
  • Multi-modal interaction

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Development of user-adaptive value system of learning function using interactive EC. / Suga, Yuki; Sugano, Shigeki; Ikuma, Yoshinori; Ogata, Tetsuya.

IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 17 1 PART 1. ed. 2008.

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

Suga, Y, Sugano, S, Ikuma, Y & Ogata, T 2008, Development of user-adaptive value system of learning function using interactive EC. in IFAC Proceedings Volumes (IFAC-PapersOnline). 1 PART 1 edn, vol. 17, 17th World Congress, International Federation of Automatic Control, IFAC, Seoul, 08/7/6. https://doi.org/10.3182/20080706-5-KR-1001.3287
Suga, Yuki ; Sugano, Shigeki ; Ikuma, Yoshinori ; Ogata, Tetsuya. / Development of user-adaptive value system of learning function using interactive EC. IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 17 1 PART 1. ed. 2008.
@inproceedings{82bd094e5b4d402a9c5a52447f2b5bb5,
title = "Development of user-adaptive value system of learning function using interactive EC",
abstract = "Our goal is to create a user-adaptive communication-robot. We are developing a system for evaluating human-robot interactions. Although such evaluation is indispensable for learning algorithms, users' preferences are too difficult to model because they are subjective. In this study, we used the interactive evolutionary computation (IEC) to configure the value system of a learning communicationrobot. The IEC is a genetic algorithm whose fitness function is performed by the user. In our experiment, we encoded the values of sensors (reward or punishment) into genes, and subjects interacted with the learning robot. Through the interaction, the subjects evaluated the robot by touching its sensors, and the robot learned appropriate combinations between input and output. Afterward, the subjects gave their scores to the experimenter, and the scores were regarded as the fitness values of the corresponding genes. These sequences were continued until the 4 generation, and then the subjects compared three of their best genes and two of the experimenter's. We found that the user-adaptive value system is suitable for the communication-robot.",
keywords = "Intelligent robotics, Multi-modal interaction",
author = "Yuki Suga and Shigeki Sugano and Yoshinori Ikuma and Tetsuya Ogata",
year = "2008",
doi = "10.3182/20080706-5-KR-1001.3287",
language = "English",
isbn = "9783902661005",
volume = "17",
booktitle = "IFAC Proceedings Volumes (IFAC-PapersOnline)",
edition = "1 PART 1",

}

TY - GEN

T1 - Development of user-adaptive value system of learning function using interactive EC

AU - Suga, Yuki

AU - Sugano, Shigeki

AU - Ikuma, Yoshinori

AU - Ogata, Tetsuya

PY - 2008

Y1 - 2008

N2 - Our goal is to create a user-adaptive communication-robot. We are developing a system for evaluating human-robot interactions. Although such evaluation is indispensable for learning algorithms, users' preferences are too difficult to model because they are subjective. In this study, we used the interactive evolutionary computation (IEC) to configure the value system of a learning communicationrobot. The IEC is a genetic algorithm whose fitness function is performed by the user. In our experiment, we encoded the values of sensors (reward or punishment) into genes, and subjects interacted with the learning robot. Through the interaction, the subjects evaluated the robot by touching its sensors, and the robot learned appropriate combinations between input and output. Afterward, the subjects gave their scores to the experimenter, and the scores were regarded as the fitness values of the corresponding genes. These sequences were continued until the 4 generation, and then the subjects compared three of their best genes and two of the experimenter's. We found that the user-adaptive value system is suitable for the communication-robot.

AB - Our goal is to create a user-adaptive communication-robot. We are developing a system for evaluating human-robot interactions. Although such evaluation is indispensable for learning algorithms, users' preferences are too difficult to model because they are subjective. In this study, we used the interactive evolutionary computation (IEC) to configure the value system of a learning communicationrobot. The IEC is a genetic algorithm whose fitness function is performed by the user. In our experiment, we encoded the values of sensors (reward or punishment) into genes, and subjects interacted with the learning robot. Through the interaction, the subjects evaluated the robot by touching its sensors, and the robot learned appropriate combinations between input and output. Afterward, the subjects gave their scores to the experimenter, and the scores were regarded as the fitness values of the corresponding genes. These sequences were continued until the 4 generation, and then the subjects compared three of their best genes and two of the experimenter's. We found that the user-adaptive value system is suitable for the communication-robot.

KW - Intelligent robotics

KW - Multi-modal interaction

UR - http://www.scopus.com/inward/record.url?scp=79961019181&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79961019181&partnerID=8YFLogxK

U2 - 10.3182/20080706-5-KR-1001.3287

DO - 10.3182/20080706-5-KR-1001.3287

M3 - Conference contribution

AN - SCOPUS:79961019181

SN - 9783902661005

VL - 17

BT - IFAC Proceedings Volumes (IFAC-PapersOnline)

ER -