What helps where – and why? Semantic relatedness for knowledge transfer

By Marcus Rohrbach, Michael Stark, György Szarvas, Iryna Gurevych, Bernt Schiele. 2010

314 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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

Citations in the corpus

Transductive Multi-View Zero-Shot Learning (2015, background): “Note that for the auxiliary dataset, Y iS is given as each data point is labelled.”
Zero-Shot Learning—A Comprehensive Evaluation of the Good, the Bad and the Ugly (2017, background): “[31], [34], [50], [51], [52], [53], [54] which exploit other auxiliary information that reduces this annotation effort.”
Evaluation of output embeddings for fine-grained image classification (2014, background): “Such a taxonomy can be built automatically from a pre-defined ontology such as WordNet [36] [46] [1] .”
Deep Compositional Captioning: Describing Novel Object Categories without Paired Training Data (2015, methodology): “[27] uses text corpora to determine how objects are related to each other, then classifies unknown objects based on their relationship to known objects.”
Latent Embeddings for Zero-Shot Classification (2016, background): “attribute prediction [20, 30, 31], or by a direct approach, e.”
Unsupervised Domain Adaptation for Zero-Shot Learning (2015, unknown): “Extensive experiments on four challenging object and action benchmark datasets [18, 34, 32] validate the advantages of the proposed model, and demonstrate that the proposed model outperforms the state-of-the-arts naive transfer learning based models [4, 18, 1, 10, 37, 15, 19, 20, 11, 2] when applied”
Evaluating knowledge transfer and zero-shot learning in a large-scale setting (2011, methodology): “2.4 then describes how semantic relatedness is used to enable unsupervised attribute- and direct similarity-based knowledge transfer.”
Zero-Shot Recognition via Semantic Embeddings and Knowledge Graphs (2018, background): “[40, 38] mined these associations from different internet sources.”
Zero-shot recognition with unreliable attributes (2014, background): “Sup-posing the system can predict the presence of any such attribute in novel images, then adding a new category model amounts to deﬁning its attribute “signature” [8, 3, 18, 24, 19].”
Sharing features between objects and their attributes (2011, background): “Usually attribute-object associationsaremanuallyspecified, butsomeworkexplores ways to obtain them automatically [26, 6, 21 ].”
Learning beautiful (and ugly) attributes (2013, background): “These learned attributes have many applications including aesthetic quality prediction, image classification and retrieval.”
COSTA: Co-Occurrence Statistics for Zero-Shot Classification (2014, methodology): “Following [21], we use the hit counts estimations of the Yahoo web search, Yahoo image search and Flickr-tag search engines to estimate to co-occurrences.”
Augmented Attribute Representations (2012, background): “It appears wasteful to use zero-shot learning in this case, but it has also been observed previously that discriminative techniques tend to fail in this regime [12,13], unless specific transfer learning techniques can be incorporated [14,15].”