English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep representation learning has become one of the most widely adoptedapproaches for visual search, recommendation, and identification. Retrieval ofsuch representations from a large database is however computationallychallenging. Approximate methods based on learning compact representations,have been widely explored for this problem, such as locality sensitive hashing,product quantization, and PCA. In this work, in contrast to learning compactrepresentations, we propose to learn high dimensional and sparserepresentations that have similar representational capacity as dense embeddingswhile being more efficient due to sparse matrix multiplication operations whichcan be much faster than dense multiplication. Following the key insight thatthe number of operations decreases quadratically with the sparsity ofembeddings provided the non-zero entries are distributed uniformly acrossdimensions, we propose a novel approach to learn such distributed sparseembeddings via the use of a carefully constructed regularization function thatdirectly minimizes a continuous relaxation of the number of floating-pointoperations (FLOPs) incurred during retrieval. Our experiments show that ourapproach is competitive to the other baselines and yields a similar or betterspeed-vs-accuracy tradeoff on practical datasets.

Minimizing FLOPs to Learn Efficient Sparse Representations