Efficient Optimization for Average Precision SVM - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Efficient Optimization for Average Precision SVM


The accuracy of information retrieval systems is often measured using average precision (AP). Given a set of positive (relevant) and negative (non-relevant) samples, the parameters of a retrieval system can be estimated using the AP-SVM framework, which minimizes a regularized convex upper bound on the empirical AP loss. However, the high computational complexity of loss-augmented inference, which is required for learning an AP-SVM, prohibits its use on large training datasets. To alleviate this deficiency, we propose three complementary approaches. The first approach guarantees an asymptotic decrease in the computational complexity of loss-augmented inference by exploiting the problem structure. The second approach takes advantage of the fact that we do not require a full ranking during loss-augmented inference. This helps us to avoid the expensive step of sorting the negative samples according to their individual scores. The third approach approximates the AP loss over all samples by the AP loss over difficult samples (for example, those that are incorrectly classified by a binary SVM), while ensuring the correct classification of the remaining samples. Using the PASCAL VOC action classification dataset, we show that our approaches provide significant speed-ups during training without degrading the test accuracy of AP-SVM.
Fichier principal
Vignette du fichier
MJK-NIPS2014.pdf (259.49 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01069917 , version 1 (30-09-2014)


  • HAL Id : hal-01069917 , version 1


Pritish Mohapatra, C. V. Jawahar, M. Pawan Kumar. Efficient Optimization for Average Precision SVM. NIPS - Advances in Neural Information Processing Systems, 2014, Montreal, Canada. ⟨hal-01069917⟩
141 View
202 Download


Gmail Facebook Twitter LinkedIn More