Abstract
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the International Conference on Computer Vision (ICCV), 2017 |
| Publisher | IEEE |
| Pages | 3553-3561 |
| Number of pages | 9 |
| Publication status | Published - 22 Oct 2017 |
Fingerprint
Cite this
Predictor Combination at Test Time. / Kim, Kwang In; Tompkin, James; Richardt, Christian.
Proceedings of the International Conference on Computer Vision (ICCV), 2017. IEEE, 2017. p. 3553-3561.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Predictor Combination at Test Time
AU - Kim, Kwang In
AU - Tompkin, James
AU - Richardt, Christian
PY - 2017/10/22
Y1 - 2017/10/22
N2 - We present an algorithm for test-time combination of a set of reference predictors with unknown parametric forms. Existing multi-task and transfer learning algorithms focus on training-time transfer and combination, where the parametric forms of predictors are known and shared. However, when the parametric form of a predictor is unknown, e.g., for a human predictor or a predictor in a precompiled library, existing algorithms are not applicable. Instead, we empirically evaluate predictors on sampled data points to measure distances between different predictors. This embeds the set of reference predictors into a Riemannian manifold, upon which we perform manifold denoising to obtain the refined predictor. This allows our approach to make no assumptions about the underlying predictor forms. Our test-time combination algorithm equals or outperforms existing multi-task and transfer learning algorithms on challenging real-world datasets, without introducing specific model assumptions.
AB - We present an algorithm for test-time combination of a set of reference predictors with unknown parametric forms. Existing multi-task and transfer learning algorithms focus on training-time transfer and combination, where the parametric forms of predictors are known and shared. However, when the parametric form of a predictor is unknown, e.g., for a human predictor or a predictor in a precompiled library, existing algorithms are not applicable. Instead, we empirically evaluate predictors on sampled data points to measure distances between different predictors. This embeds the set of reference predictors into a Riemannian manifold, upon which we perform manifold denoising to obtain the refined predictor. This allows our approach to make no assumptions about the underlying predictor forms. Our test-time combination algorithm equals or outperforms existing multi-task and transfer learning algorithms on challenging real-world datasets, without introducing specific model assumptions.
UR - http://openaccess.thecvf.com/content_iccv_2017/html/Kim_Predictor_Combination_at_ICCV_2017_paper.html
M3 - Conference contribution
SP - 3553
EP - 3561
BT - Proceedings of the International Conference on Computer Vision (ICCV), 2017
PB - IEEE
ER -