Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288093006> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4288093006 abstract "Gaussian Processes (GPs) are known to provide accurate predictions and uncertainty estimates even with small amounts of labeled data by capturing similarity between data points through their kernel function. However traditional GP kernels are not very effective at capturing similarity between high dimensional data points. Neural networks can be used to learn good representations that encode intricate structures in high dimensional data, and can be used as inputs to the GP kernel. However the huge data requirement of neural networks makes this approach ineffective in small data settings. To solves the conflicting problems of representation learning and data efficiency, we propose to learn deep kernels on probabilistic embeddings by using a probabilistic neural network. Our approach maps high-dimensional data to a probability distribution in a low dimensional subspace and then computes a kernel between these distributions to capture similarity. To enable end-to-end learning, we derive a functional gradient descent procedure for training the model. Experiments on a variety of datasets show that our approach outperforms the state-of-the-art in GP kernel learning in both supervised and semi-supervised settings. We also extend our approach to other small-data paradigms such as few-shot classification where it outperforms previous approaches on mini-Imagenet and CUB datasets." @default.
- W4288093006 created "2022-07-28" @default.
- W4288093006 creator A5018843444 @default.
- W4288093006 creator A5021638015 @default.
- W4288093006 creator A5041470575 @default.
- W4288093006 creator A5046601788 @default.
- W4288093006 creator A5067441201 @default.
- W4288093006 date "2019-10-13" @default.
- W4288093006 modified "2023-10-16" @default.
- W4288093006 title "Deep Kernels with Probabilistic Embeddings for Small-Data Learning" @default.
- W4288093006 doi "https://doi.org/10.48550/arxiv.1910.05858" @default.
- W4288093006 hasPublicationYear "2019" @default.
- W4288093006 type Work @default.
- W4288093006 citedByCount "0" @default.
- W4288093006 crossrefType "posted-content" @default.
- W4288093006 hasAuthorship W4288093006A5018843444 @default.
- W4288093006 hasAuthorship W4288093006A5021638015 @default.
- W4288093006 hasAuthorship W4288093006A5041470575 @default.
- W4288093006 hasAuthorship W4288093006A5046601788 @default.
- W4288093006 hasAuthorship W4288093006A5067441201 @default.
- W4288093006 hasBestOaLocation W42880930061 @default.
- W4288093006 hasConcept C103278499 @default.
- W4288093006 hasConcept C114614502 @default.
- W4288093006 hasConcept C115961682 @default.
- W4288093006 hasConcept C119857082 @default.
- W4288093006 hasConcept C122280245 @default.
- W4288093006 hasConcept C12267149 @default.
- W4288093006 hasConcept C153180895 @default.
- W4288093006 hasConcept C153258448 @default.
- W4288093006 hasConcept C154945302 @default.
- W4288093006 hasConcept C17744445 @default.
- W4288093006 hasConcept C199539241 @default.
- W4288093006 hasConcept C21080849 @default.
- W4288093006 hasConcept C2776359362 @default.
- W4288093006 hasConcept C32834561 @default.
- W4288093006 hasConcept C33923547 @default.
- W4288093006 hasConcept C41008148 @default.
- W4288093006 hasConcept C49937458 @default.
- W4288093006 hasConcept C50644808 @default.
- W4288093006 hasConcept C59404180 @default.
- W4288093006 hasConcept C74193536 @default.
- W4288093006 hasConcept C94625758 @default.
- W4288093006 hasConceptScore W4288093006C103278499 @default.
- W4288093006 hasConceptScore W4288093006C114614502 @default.
- W4288093006 hasConceptScore W4288093006C115961682 @default.
- W4288093006 hasConceptScore W4288093006C119857082 @default.
- W4288093006 hasConceptScore W4288093006C122280245 @default.
- W4288093006 hasConceptScore W4288093006C12267149 @default.
- W4288093006 hasConceptScore W4288093006C153180895 @default.
- W4288093006 hasConceptScore W4288093006C153258448 @default.
- W4288093006 hasConceptScore W4288093006C154945302 @default.
- W4288093006 hasConceptScore W4288093006C17744445 @default.
- W4288093006 hasConceptScore W4288093006C199539241 @default.
- W4288093006 hasConceptScore W4288093006C21080849 @default.
- W4288093006 hasConceptScore W4288093006C2776359362 @default.
- W4288093006 hasConceptScore W4288093006C32834561 @default.
- W4288093006 hasConceptScore W4288093006C33923547 @default.
- W4288093006 hasConceptScore W4288093006C41008148 @default.
- W4288093006 hasConceptScore W4288093006C49937458 @default.
- W4288093006 hasConceptScore W4288093006C50644808 @default.
- W4288093006 hasConceptScore W4288093006C59404180 @default.
- W4288093006 hasConceptScore W4288093006C74193536 @default.
- W4288093006 hasConceptScore W4288093006C94625758 @default.
- W4288093006 hasLocation W42880930061 @default.
- W4288093006 hasOpenAccess W4288093006 @default.
- W4288093006 hasPrimaryLocation W42880930061 @default.
- W4288093006 hasRelatedWork W1591413801 @default.
- W4288093006 hasRelatedWork W1996930579 @default.
- W4288093006 hasRelatedWork W2067378131 @default.
- W4288093006 hasRelatedWork W2068555361 @default.
- W4288093006 hasRelatedWork W2081053296 @default.
- W4288093006 hasRelatedWork W2106922626 @default.
- W4288093006 hasRelatedWork W2125244435 @default.
- W4288093006 hasRelatedWork W3184787677 @default.
- W4288093006 hasRelatedWork W3212130925 @default.
- W4288093006 hasRelatedWork W2520220765 @default.
- W4288093006 isParatext "false" @default.
- W4288093006 isRetracted "false" @default.
- W4288093006 workType "article" @default.