A Non-parametric Conformity-Based Test for Transfer Decisions

S. Zhou; E. N. Smirnov; G. Schoenmakers; R. Peeters; K. Driessens

doi:10.1109/ICTAI.2015.96

A Non-parametric Conformity-Based Test for Transfer Decisions

S. Zhou^*, E. N. Smirnov, G. Schoenmakers, R. Peeters, K. Driessens

^*Corresponding author for this work

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

6 Downloads (Pure)

Abstract

This paper proposes a new non-parametric test to decide whether to transfer from source data to target data in order to improve the performance of predictive models on target domains. The test is based on the conformity framework. It statistically tests whether the target data and source data have been generated from the target distribution under the exchangeability assumption. The source data is transferred if and only if the test is positive. The experiments show that the test is better at deciding on instance transfer than existing methods.

Original language	English
Title of host publication	Proceedings of the 27th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2015)
Place of Publication	Vietri sul Mare, Italy
Pages	628-635
Number of pages	8
DOIs	https://doi.org/10.1109/ICTAI.2015.96
Publication status	Published - 2015

Access to Document

10.1109/ICTAI.2015.96

Cite this

@inproceedings{14d8d043b3ed4fc4925b0d411cbafd24,

title = "A Non-parametric Conformity-Based Test for Transfer Decisions",

abstract = "This paper proposes a new non-parametric test to decide whether to transfer from source data to target data in order to improve the performance of predictive models on target domains. The test is based on the conformity framework. It statistically tests whether the target data and source data have been generated from the target distribution under the exchangeability assumption. The source data is transferred if and only if the test is positive. The experiments show that the test is better at deciding on instance transfer than existing methods.",

author = "S. Zhou and Smirnov, {E. N.} and G. Schoenmakers and R. Peeters and K. Driessens",

year = "2015",

doi = "10.1109/ICTAI.2015.96",

language = "English",

pages = "628--635",

booktitle = "Proceedings of the 27th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2015)",

}

A Non-parametric Conformity-Based Test for Transfer Decisions. / Zhou, S.; Smirnov, E. N.; Schoenmakers, G. et al.
Proceedings of the 27th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2015). Vietri sul Mare, Italy, 2015. p. 628-635.

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

TY - GEN

T1 - A Non-parametric Conformity-Based Test for Transfer Decisions

AU - Zhou, S.

AU - Smirnov, E. N.

AU - Schoenmakers, G.

AU - Peeters, R.

AU - Driessens, K.

PY - 2015

Y1 - 2015

N2 - This paper proposes a new non-parametric test to decide whether to transfer from source data to target data in order to improve the performance of predictive models on target domains. The test is based on the conformity framework. It statistically tests whether the target data and source data have been generated from the target distribution under the exchangeability assumption. The source data is transferred if and only if the test is positive. The experiments show that the test is better at deciding on instance transfer than existing methods.

AB - This paper proposes a new non-parametric test to decide whether to transfer from source data to target data in order to improve the performance of predictive models on target domains. The test is based on the conformity framework. It statistically tests whether the target data and source data have been generated from the target distribution under the exchangeability assumption. The source data is transferred if and only if the test is positive. The experiments show that the test is better at deciding on instance transfer than existing methods.

U2 - 10.1109/ICTAI.2015.96

DO - 10.1109/ICTAI.2015.96

M3 - Conference article in proceeding

SP - 628

EP - 635

BT - Proceedings of the 27th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2015)

CY - Vietri sul Mare, Italy

ER -