Explainable Search: An Exploratory Study in SameGame

Chiara F. Sironi; Anna Wilbik; Mark H. M. Winands

doi:10.1109/CoG57401.2023.10333232

Explainable Search: An Exploratory Study in SameGame

Chiara F. Sironi, Anna Wilbik, Mark H. M. Winands

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceeding › Academic › peer-review

10 Downloads (Pure)

Abstract

The field of Explainable Artificial Intelligence has gained popularity in recent years, due to the need for users to understand AI-made decisions, in order to increase their trust in the AI system. However, not much work has been performed on explaining recommendations made by search algorithms, which do not focus on single decisions, but on complex plans of action. This paper investigates promising directions for research in Explainable Search (XS), by evaluating with a user study different types of explanations for a search-based algorithm. Preliminary results suggest that users prefer explanations generated using context-based features, which are not only based on the current state of the problem, but are extracted from different parts of the tree generated by the search algorithm.

Original language	English
Title of host publication	2023 IEEE Conference on Games (CoG)
Publisher	IEEE
Pages	1-4
ISBN (Electronic)	979-8-3503-2277-4
ISBN (Print)	979-8-3503-2278-1
DOIs	https://doi.org/10.1109/CoG57401.2023.10333232
Publication status	Published - 21 Aug 2023
Event	2023 IEEE Conference on Games (CoG) - Boston, United States Duration: 21 Aug 2023 → 24 Aug 2023 https://2023.ieee-cog.org/

Conference

Conference	2023 IEEE Conference on Games (CoG)
Abbreviated title	IEEE CoG
Country/Territory	United States
City	Boston
Period	21/08/23 → 24/08/23
Internet address	https://2023.ieee-cog.org/

Access to Document

10.1109/CoG57401.2023.10333232

Full TextFinal published version, 251 KBLicence: Taverne

https://ieeexplore.ieee.org/document/10333232/

Cite this

@inproceedings{8740f5db3e794899af5af50eeca87dbd,

title = "Explainable Search: An Exploratory Study in SameGame",

abstract = "The field of Explainable Artificial Intelligence has gained popularity in recent years, due to the need for users to understand AI-made decisions, in order to increase their trust in the AI system. However, not much work has been performed on explaining recommendations made by search algorithms, which do not focus on single decisions, but on complex plans of action. This paper investigates promising directions for research in Explainable Search (XS), by evaluating with a user study different types of explanations for a search-based algorithm. Preliminary results suggest that users prefer explanations generated using context-based features, which are not only based on the current state of the problem, but are extracted from different parts of the tree generated by the search algorithm.",

author = "Sironi, {Chiara F.} and Anna Wilbik and Winands, {Mark H. M.}",

year = "2023",

month = aug,

day = "21",

doi = "10.1109/CoG57401.2023.10333232",

language = "English",

isbn = "979-8-3503-2278-1",

pages = "1--4",

booktitle = "2023 IEEE Conference on Games (CoG)",

publisher = "IEEE",

address = "United States",

note = "2023 IEEE Conference on Games (CoG), IEEE CoG ; Conference date: 21-08-2023 Through 24-08-2023",

url = "https://2023.ieee-cog.org/",

}

TY - GEN

T1 - Explainable Search: An Exploratory Study in SameGame

AU - Sironi, Chiara F.

AU - Wilbik, Anna

AU - Winands, Mark H. M.

PY - 2023/8/21

Y1 - 2023/8/21

N2 - The field of Explainable Artificial Intelligence has gained popularity in recent years, due to the need for users to understand AI-made decisions, in order to increase their trust in the AI system. However, not much work has been performed on explaining recommendations made by search algorithms, which do not focus on single decisions, but on complex plans of action. This paper investigates promising directions for research in Explainable Search (XS), by evaluating with a user study different types of explanations for a search-based algorithm. Preliminary results suggest that users prefer explanations generated using context-based features, which are not only based on the current state of the problem, but are extracted from different parts of the tree generated by the search algorithm.

AB - The field of Explainable Artificial Intelligence has gained popularity in recent years, due to the need for users to understand AI-made decisions, in order to increase their trust in the AI system. However, not much work has been performed on explaining recommendations made by search algorithms, which do not focus on single decisions, but on complex plans of action. This paper investigates promising directions for research in Explainable Search (XS), by evaluating with a user study different types of explanations for a search-based algorithm. Preliminary results suggest that users prefer explanations generated using context-based features, which are not only based on the current state of the problem, but are extracted from different parts of the tree generated by the search algorithm.

U2 - 10.1109/CoG57401.2023.10333232

DO - 10.1109/CoG57401.2023.10333232

M3 - Conference article in proceeding

SN - 979-8-3503-2278-1

SP - 1

EP - 4

BT - 2023 IEEE Conference on Games (CoG)

PB - IEEE

T2 - 2023 IEEE Conference on Games (CoG)

Y2 - 21 August 2023 through 24 August 2023

ER -