GENERALIZABLE EMBEDDINGS WITH CROSS-BATCH METRIC LEARNING

Date

2023-10-08

Author

GÜRBÜZ, YETİ ZİYA
Alatan, Abdullah Aydın

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

93
views

0
downloads

Cite This

Global average pooling (GAP) is a popular component in deep metric learning (DML) for aggregating features. Its effectiveness is often attributed to treating each feature vector as a distinct semantic entity and GAP as a combination of them. Albeit substantiated, such an explanation's algorithmic implications to learn generalizable entities to represent unseen classes, a crucial DML goal, remain unclear. To address this, we formulate GAP as a convex combination of learnable prototypes. We then show that the prototype learning can be expressed as a recursive process fitting a linear predictor to a batch of samples. Building on that perspective, we consider two batches of disjoint classes at each iteration and regularize the learning by expressing the samples of a batch with the prototypes that are fitted to the other batch. We validate our approach on 4 popular DML benchmarks.

URI

https://ieeexplore.ieee.org/xpl/conhome/10221937/proceeding
https://hdl.handle.net/11511/111780

Collections

Department of Electrical and Electronics Engineering, Conference / Seminar