We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Neighborhood Sensitive Mapping for Zero-Shot Classification using Independently Learned Semantic Embeddings

Abstract: In a traditional setting, classifiers are trained to approximate a target function $f:X \rightarrow Y$ where at least a sample for each $y \in Y$ is presented to the training algorithm. In a zero-shot setting we have a subset of the labels $\hat{Y} \subset Y$ for which we do not observe any corresponding training instance. Still, the function $f$ that we train must be able to correctly assign labels also on $\hat{Y}$. In practice, zero-shot problems are very important especially when the label set is large and the cost of editorially label samples for all possible values in the label set might be prohibitively high. Most recent approaches to zero-shot learning are based on finding and exploiting relationships between labels using semantic embeddings. We show in this paper that semantic embeddings, despite being very good at capturing relationships between labels, are not very good at capturing the relationships among labels in a data-dependent manner. For this reason, we propose a novel two-step process for learning a zero-shot classifier. In the first step, we learn what we call a \emph{property embedding space} capturing the "\emph{learnable}" features of the label set. Then, we exploit the learned properties in order to reduce the generalization error for a linear nearest neighbor-based classifier.
Comments: ACML Workshop on Learning on Big Data, 2016
Subjects: Machine Learning (cs.LG)
Cite as: arXiv:1605.08242 [cs.LG]
  (or arXiv:1605.08242v3 [cs.LG] for this version)

Submission history

From: Gaurav Singh [view email]
[v1] Thu, 26 May 2016 11:53:26 GMT (192kb)
[v2] Tue, 18 Dec 2018 15:23:08 GMT (192kb)
[v3] Wed, 19 Aug 2020 13:40:11 GMT (198kb)

Link back to: arXiv, form interface, contact.