Risk-Sensitive Learning via Minimization of Empirical Conditional Value-at-Risk

Hisashi KASHIMA

doi:10.1093/ietisy/e90-d.12.2043

Risk-Sensitive Learning via Minimization of Empirical Conditional Value-at-Risk

Hisashi KASHIMA

Full Text Views

0

Share
Cite this

Summary :

We extend the framework of cost-sensitive classification to mitigate risks of huge costs occurring with low probabilities, and propose an algorithm that achieves this goal. Instead of minimizing the expected cost commonly used in cost-sensitive learning, our algorithm minimizes conditional value-at-risk, also known as expected shortfall, which is considered a good risk metric in the area of financial engineering. The proposed algorithm is a general meta-learning algorithm that can exploit existing example-dependent cost-sensitive learning algorithms, and is capable of dealing with not only alternative actions in ordinary classification tasks, but also allocative actions in resource-allocation type tasks. Experiments on tasks with example-dependent costs show promising results.

Publication: IEICE TRANSACTIONS on Information Vol.E90-D No.12 pp.2043-2052

Publication Date: 2007/12/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1093/ietisy/e90-d.12.2043

Type of Manuscript: PAPER

Category: Artificial Intelligence and Cognitive Science

Cite this

Copy

Hisashi KASHIMA, "Risk-Sensitive Learning via Minimization of Empirical Conditional Value-at-Risk" in IEICE TRANSACTIONS on Information, vol. E90-D, no. 12, pp. 2043-2052, December 2007, doi: 10.1093/ietisy/e90-d.12.2043.
Abstract: We extend the framework of cost-sensitive classification to mitigate risks of huge costs occurring with low probabilities, and propose an algorithm that achieves this goal. Instead of minimizing the expected cost commonly used in cost-sensitive learning, our algorithm minimizes conditional value-at-risk, also known as expected shortfall, which is considered a good risk metric in the area of financial engineering. The proposed algorithm is a general meta-learning algorithm that can exploit existing example-dependent cost-sensitive learning algorithms, and is capable of dealing with not only alternative actions in ordinary classification tasks, but also allocative actions in resource-allocation type tasks. Experiments on tasks with example-dependent costs show promising results.
URL: https://globals.ieice.org/en_transactions/information/10.1093/ietisy/e90-d.12.2043/_p

Copy

@ARTICLE{e90-d_12_2043,
author={Hisashi KASHIMA, },
journal={IEICE TRANSACTIONS on Information},
title={Risk-Sensitive Learning via Minimization of Empirical Conditional Value-at-Risk},
year={2007},
volume={E90-D},
number={12},
pages={2043-2052},
abstract={We extend the framework of cost-sensitive classification to mitigate risks of huge costs occurring with low probabilities, and propose an algorithm that achieves this goal. Instead of minimizing the expected cost commonly used in cost-sensitive learning, our algorithm minimizes conditional value-at-risk, also known as expected shortfall, which is considered a good risk metric in the area of financial engineering. The proposed algorithm is a general meta-learning algorithm that can exploit existing example-dependent cost-sensitive learning algorithms, and is capable of dealing with not only alternative actions in ordinary classification tasks, but also allocative actions in resource-allocation type tasks. Experiments on tasks with example-dependent costs show promising results.},
keywords={},
doi={10.1093/ietisy/e90-d.12.2043},
ISSN={1745-1361},
month={December},}

Copy

TY - JOUR
TI - Risk-Sensitive Learning via Minimization of Empirical Conditional Value-at-Risk
T2 - IEICE TRANSACTIONS on Information
SP - 2043
EP - 2052
AU - Hisashi KASHIMA
PY - 2007
DO - 10.1093/ietisy/e90-d.12.2043
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E90-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2007
AB - We extend the framework of cost-sensitive classification to mitigate risks of huge costs occurring with low probabilities, and propose an algorithm that achieves this goal. Instead of minimizing the expected cost commonly used in cost-sensitive learning, our algorithm minimizes conditional value-at-risk, also known as expected shortfall, which is considered a good risk metric in the area of financial engineering. The proposed algorithm is a general meta-learning algorithm that can exploit existing example-dependent cost-sensitive learning algorithms, and is capable of dealing with not only alternative actions in ordinary classification tasks, but also allocative actions in resource-allocation type tasks. Experiments on tasks with example-dependent costs show promising results.
ER -