Matches in SemOpenAlex for { <https://semopenalex.org/work/W2898885327> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2898885327 endingPage "15" @default.
- W2898885327 startingPage "1" @default.
- W2898885327 abstract "We investigate the ways in which a machine learning architecture known as Reservoir Computing learns concepts such as “similar” and “different” and other relationships between image pairs and generalizes these concepts to previously unseen classes of data. We present two Reservoir Computing architectures, which loosely resemble neural dynamics, and show that a Reservoir Computer (RC) trained to identify relationships between image pairs drawn from a subset of training classes generalizes the learned relationships to substantially different classes unseen during training. We demonstrate our results on the simple MNIST handwritten digit database as well as a database of depth maps of visual scenes in videos taken from a moving camera. We consider image pair relationships such as images from the same class; images from the same class with one image superposed with noise, rotated 90°, blurred, or scaled; images from different classes. We observe that the reservoir acts as a nonlinear filter projecting the input into a higher dimensional space in which the relationships are separable; i.e., the reservoir system state trajectories display different dynamical patterns that reflect the corresponding input pair relationships. Thus, as opposed to training in the entire high-dimensional reservoir space, the RC only needs to learns characteristic features of these dynamical patterns, allowing it to perform well with very few training examples compared with conventional machine learning feed-forward techniques such as deep learning. In generalization tasks, we observe that RCs perform significantly better than state-of-the-art, feed-forward, pair-based architectures such as convolutional and deep Siamese Neural Networks (SNNs). We also show that RCs can not only generalize relationships, but also generalize combinations of relationships, providing robust and effective image pair classification. Our work helps bridge the gap between explainable machine learning with small datasets and biologically inspired analogy-based learning, pointing to new directions in the investigation of learning processes." @default.
- W2898885327 created "2018-11-09" @default.
- W2898885327 creator A5002601106 @default.
- W2898885327 creator A5017906200 @default.
- W2898885327 creator A5036912867 @default.
- W2898885327 date "2018-11-01" @default.
- W2898885327 modified "2023-10-16" @default.
- W2898885327 title "Similarity Learning and Generalization with Limited Data: A Reservoir Computing Approach" @default.
- W2898885327 cites W1546307650 @default.
- W2898885327 cites W2008284899 @default.
- W2898885327 cites W2029967456 @default.
- W2898885327 cites W2036202071 @default.
- W2898885327 cites W2040640391 @default.
- W2898885327 cites W2095396608 @default.
- W2898885327 cites W2103179919 @default.
- W2898885327 cites W2109840317 @default.
- W2898885327 cites W2127918219 @default.
- W2898885327 cites W2128084896 @default.
- W2898885327 cites W2137944749 @default.
- W2898885327 cites W2158375962 @default.
- W2898885327 cites W2171865010 @default.
- W2898885327 cites W2321627895 @default.
- W2898885327 cites W2618530766 @default.
- W2898885327 cites W2765128778 @default.
- W2898885327 cites W2772813843 @default.
- W2898885327 cites W2919115771 @default.
- W2898885327 cites W2963924008 @default.
- W2898885327 doi "https://doi.org/10.1155/2018/6953836" @default.
- W2898885327 hasPublicationYear "2018" @default.
- W2898885327 type Work @default.
- W2898885327 sameAs 2898885327 @default.
- W2898885327 citedByCount "12" @default.
- W2898885327 countsByYear W28988853272019 @default.
- W2898885327 countsByYear W28988853272020 @default.
- W2898885327 countsByYear W28988853272021 @default.
- W2898885327 countsByYear W28988853272022 @default.
- W2898885327 countsByYear W28988853272023 @default.
- W2898885327 crossrefType "journal-article" @default.
- W2898885327 hasAuthorship W2898885327A5002601106 @default.
- W2898885327 hasAuthorship W2898885327A5017906200 @default.
- W2898885327 hasAuthorship W2898885327A5036912867 @default.
- W2898885327 hasBestOaLocation W28988853271 @default.
- W2898885327 hasConcept C103278499 @default.
- W2898885327 hasConcept C108583219 @default.
- W2898885327 hasConcept C115961682 @default.
- W2898885327 hasConcept C134306372 @default.
- W2898885327 hasConcept C135796866 @default.
- W2898885327 hasConcept C147168706 @default.
- W2898885327 hasConcept C153180895 @default.
- W2898885327 hasConcept C154945302 @default.
- W2898885327 hasConcept C177148314 @default.
- W2898885327 hasConcept C190502265 @default.
- W2898885327 hasConcept C2777212361 @default.
- W2898885327 hasConcept C33923547 @default.
- W2898885327 hasConcept C41008148 @default.
- W2898885327 hasConcept C50644808 @default.
- W2898885327 hasConcept C81363708 @default.
- W2898885327 hasConcept C99498987 @default.
- W2898885327 hasConceptScore W2898885327C103278499 @default.
- W2898885327 hasConceptScore W2898885327C108583219 @default.
- W2898885327 hasConceptScore W2898885327C115961682 @default.
- W2898885327 hasConceptScore W2898885327C134306372 @default.
- W2898885327 hasConceptScore W2898885327C135796866 @default.
- W2898885327 hasConceptScore W2898885327C147168706 @default.
- W2898885327 hasConceptScore W2898885327C153180895 @default.
- W2898885327 hasConceptScore W2898885327C154945302 @default.
- W2898885327 hasConceptScore W2898885327C177148314 @default.
- W2898885327 hasConceptScore W2898885327C190502265 @default.
- W2898885327 hasConceptScore W2898885327C2777212361 @default.
- W2898885327 hasConceptScore W2898885327C33923547 @default.
- W2898885327 hasConceptScore W2898885327C41008148 @default.
- W2898885327 hasConceptScore W2898885327C50644808 @default.
- W2898885327 hasConceptScore W2898885327C81363708 @default.
- W2898885327 hasConceptScore W2898885327C99498987 @default.
- W2898885327 hasFunder F4320310578 @default.
- W2898885327 hasLocation W28988853271 @default.
- W2898885327 hasLocation W28988853272 @default.
- W2898885327 hasOpenAccess W2898885327 @default.
- W2898885327 hasPrimaryLocation W28988853271 @default.
- W2898885327 hasRelatedWork W2285788670 @default.
- W2898885327 hasRelatedWork W2732542196 @default.
- W2898885327 hasRelatedWork W2738221750 @default.
- W2898885327 hasRelatedWork W2900794075 @default.
- W2898885327 hasRelatedWork W2909060151 @default.
- W2898885327 hasRelatedWork W2947175736 @default.
- W2898885327 hasRelatedWork W3156786002 @default.
- W2898885327 hasRelatedWork W4309224979 @default.
- W2898885327 hasRelatedWork W4317374280 @default.
- W2898885327 hasRelatedWork W564581980 @default.
- W2898885327 hasVolume "2018" @default.
- W2898885327 isParatext "false" @default.
- W2898885327 isRetracted "false" @default.
- W2898885327 magId "2898885327" @default.
- W2898885327 workType "article" @default.