University of Sussex
Browse

File(s) under permanent embargo

Beyond dataset bias: multi-task unaligned shared knowledge transfer

conference contribution
posted on 2023-06-08, 16:45 authored by Tatiana Tommasi, Novi QuadriantoNovi Quadrianto, Barbara Caputo, Christoph H Lampert
Many visual datasets are traditionally used to analyze the performance of different learning techniques. The evaluation is usually done within each dataset, therefore it is questionable if such results are a reliable indicator of true generalization ability. We propose here an algorithm to exploit the existing data resources when learning on a new multiclass problem. Our main idea is to identify an image representation that decomposes orthogonally into two subspaces: a part specific to each dataset, and a part generic to, and therefore shared between, all the considered source sets. This allows us to use the generic representation as un-biased reference knowledge for a novel classification task. By casting the method in the multi-view setting, we also make it possible to use different features for different databases. We call the algorithm MUST, Multitask Unaligned Shared knowledge Transfer. Through extensive experiments on five public datasets, we show that MUST consistently improves the cross-datasets generalization performance.

History

Publication status

  • Published

File Version

  • Published version

Journal

Proceedings of Computer vision - ACCV 2012: 11th Asian Conference on Computer Vision; Daejeon, Korea; 5-9 November 2012

ISSN

0302-9743

Publisher

Springer Verlag

Issue

7724

Page range

1-15

Pages

821.0

Event name

11th Asian conference on computer vision (ACCV)

Event location

Daejeon, Korea

Event type

conference

Event date

5-9 November, 2012

Book title

Computer vision – ACCV 2012: 11th Asian conference on computer vision, Daejeon, Korea, November 5-9, 2012, revised selected papers, Part I

Place of publication

Berlin; New York

ISBN

9783642373305

Series

Lecture notes in computer science

Department affiliated with

  • Informatics Publications

Full text available

  • No

Peer reviewed?

  • Yes

Editors

Kyoung Mu Lee, Zhanyi Hu, James M Rehg, Yasuyuki Matsushita

Legacy Posted Date

2014-02-25

First Compliant Deposit (FCD) Date

2014-02-25

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC