Label-Embedding for Attribute-Based Classification

By Zeynep Akata, Florent Perronnin, Zaïd Harchaoui, Cordelia Schmid. 2013

699 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Representation & Self-Supervised Learning

Citations in the corpus

Prototypical Networks for Few-shot Learning (2017, background): “For prototypical networks, we found that it is usually best to train and test with the same “shot” number.”
Transductive Multi-View Zero-Shot Learning (2015, methodology): “They can be roughly categorised according to the semantic representation(s) used: DAP and IAP ([27], [28]), M2LATM [15], ALE [1], [36] and [50] use attributes only; HLE/AHLE [1] and Mo/Ma/O/D [38] use both attributes and linguistic knowledge bases (same as us); [54] uses attribute and some…”
Subspace Regularizers for Few-Shot Class Incremental Learning (2021, background): “…et al., 2021; Reed et al., 2016; Yan et al., 2021)—particularly in the zero-shot learning and generalized zero-shot learning where side information is the only source of information about the novel class (Chang et al., 2008; Larochelle et al., 2008; Akata et al., 2013; Pourpanah et al., 2020).”
Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly (2017, methodology): “Different flavors of zero-shot learning methods that we evaluate in this work are linear [7], [8], [9], [10] and nonlinear [11], [12] compatibility learning frameworks which have dominated the zero-shot learning literature in the past few years whereas an orthogonal direction is learning independent”
Zero-Shot Learning — The Good, the Bad and the Ugly (2017, methodology): “we term them as standard splits (SS), for SUN from [21] and CUB from [2] shows us that 1 CUB test class out of 50 (Indigo Bunting),”
An embarrassingly simple approach to zero-shot learning (2015, methodology): “One interpretation of thismodel is provided in [1].”
Evaluation of output embeddings for fine-grained image classification (2014, methodology): “Within the label embedding framework, ALE [1] and DeViSe [16] use pairwise ranking objective, WSABIE [56] learns both φ(y) and W through ranking, whereas we use multiclass objective.”
Learning Adaptive Classifiers Synthesis for Generalized Few-Shot Learning (2019, background): “Zero-shot learning (ZSL) (Lampert et al. 2014;Akata et al. 2013; Xian et al. 2017; Changpinyo et al. 2020) is a popular idea for addressing learningwithout labeled data.”
Zero-Shot Object Detection (2018, neutral): “More recent approaches have used multimodal embeddings to learn a compatibility function between an image vector and class label embeddings [1,2].”
Generalizing from a Few Examples (2019, methodology): “t zero-shot learning methods to deal with FSL problems by the a supervised data in D train. These methods either use the few-shot in D to fine-tune the parameter learned by zero-shot learning methods [1, 50], or force the embedding or model learned by multiple modalities to match on a shared space [”
Label-Embedding for Image Classification (2015, methodology): “As opposed to [1], we do not apply PQcompression which explains why we report better results in the current work (e.”
Few-Shot Learning via Embedding Adaptation With Set-to-Set Functions (2018, background): “Another line of approach has focused on learning generalizable instance embeddings [1, 4, 5, 13, 17, 26, 37, 40, 43] and uses those embeddings on simple classifiers such as nearest neighbor rules.”
Zero-Shot Learning via Semantic Similarity Embedding (2015, methodology): “This leads to complex scoring functions, which cannot be simplified to linear or bilinear forms as in [1, 2].”
MSplit LBI: Realizing Feature Selection and Dense Estimation Simultaneously in Few-shot and Zero-shot Learning (2018, methodology): “Our embedding model is compared against the state-of-the-art methods, including DAP/IAP (Lampert et al. (2014)), SJE (Akata et al. (2015)), SC_struct (Changpinyo et al. (2016)), LatEm (Xian et al. (2016)), LEESD (Ding et al. (2017)), SS-Voc (Fu and Sigal (2016)), DCL (Guo et al. (2017)), JLSE (Zhang”
Learning Embedding Adaptation for Few-Shot Learning (2018, background): “Another line of approach has focused on learning gener-alizable instance embeddings [1, 5, 6, 18, 28, 42, 44] and uses those embeddings on simple classiﬁers such as nearest neighbor classiﬁcation rules.”
A Survey of Zero-Shot Learning (2019, methodology): “Finally, θ ( · , · ; ς ) : R D × R M → R is a compatibility function [2, 155] that takes an instance x i and a class prototype t j as inputs and outputs a compatibility score denoting the confidence/probability about x i belonging to class c j .”
Learning Transferable Knowledge Through Embedding Spaces (2019, methodology): “Numerous ZSL algorithms have been developed to learn a mapping between the visual features and semantic attributes using a shared space [52; 53; 54; 55; 56; 19].”