.png){kind=icon}
Return
Large-margin multiple kernel ep-SVDD using Frank-Wolfe algorithm for novelty detection
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2023.110189.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Using a variable ep >= 1-norm penalty on the slacks, the recently introduced ep-norm Support Vector Data Description (ep-SVDD) method has improved the performance in novelty detection over the baseline approach, sometimes remarkably. This work extends this modelling formalism in multiple aspects. First, a large-margin extension of the ep-SVDD method is formulated to enhance generalisation capability by maximising the margin between the positive and negative samples. Second, based on the Frank-Wolfe algorithm, an efficient yet effective method with predictable accuracy is presented to optimise the convex objective function in the proposed method. Finally, it is illustrated that the proposed approach can effectively benefit from a multiple kernel learning scheme to achieve state-of-the-art performance.The proposed method is theoretically analysed using Rademacher complexities to link its classification error probability to the margin and experimentally evaluated on several datasets to demonstrate its merits against existing methods.
Keywords:
ep-SVDD
Large-margin learning
Convex optimisation
Frank-Wolfe algorithm
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W
Organization
Cited Papers
ShareSaveVideo anomaly detection and localization using motion-field shape description and homogeneity testing.png)
PATTERN RECOGNITION
PATTERN RECOGNITIONIF7.6
PREAI
ShareSaveShareSaveShareSaveClient-specific anomaly detection for face presentation attack detectionPATTERN RECOGNITION
OAAI
PATTERN RECOGNITIONIF7.6
OAAIShareSaveShareSaveMeasurement of interstrand cross-link frequency and distance between interruptions in DNA exposed to 4,5',8-trimethylpsoralen and near-ultraviolet lightBiochemistry
BiochemistryIF0
PREAI
ShareSave
ShareSave
View more