Matches in SemOpenAlex for { <https://semopenalex.org/work/W4363671961> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4363671961 abstract "Deep learning has achieved impressive performance in many domains, such as computer vision and natural language processing, but its advantage over classical shallow methods on tabular datasets remains questionable. It is especially challenging to surpass the performance of tree-like ensembles, such as XGBoost or Random Forests, on small-sized datasets (less than 1k samples). To tackle this challenge, we introduce HyperTab, a hypernetwork-based approach to solving small sample problems on tabular datasets. By combining the advantages of Random Forests and neural networks, HyperTab generates an ensemble of neural networks, where each target model is specialized to process a specific lower-dimensional view of the data. Since each view plays the role of data augmentation, we virtually increase the number of training samples while keeping the number of trainable parameters unchanged, which prevents model overfitting. We evaluated HyperTab on more than 40 tabular datasets of a varying number of samples and domains of origin, and compared its performance with shallow and deep learning models representing the current state-of-the-art. We show that HyperTab consistently outranks other methods on small data (with a statistically significant difference) and scores comparable to them on larger datasets. We make a python package with the code available to download at https://pypi.org/project/hypertab/" @default.
- W4363671961 created "2023-04-11" @default.
- W4363671961 creator A5021803359 @default.
- W4363671961 creator A5044795083 @default.
- W4363671961 creator A5089509620 @default.
- W4363671961 date "2023-04-07" @default.
- W4363671961 modified "2023-10-11" @default.
- W4363671961 title "HyperTab: Hypernetwork Approach for Deep Learning on Small Tabular Datasets" @default.
- W4363671961 doi "https://doi.org/10.48550/arxiv.2304.03543" @default.
- W4363671961 hasPublicationYear "2023" @default.
- W4363671961 type Work @default.
- W4363671961 citedByCount "0" @default.
- W4363671961 crossrefType "posted-content" @default.
- W4363671961 hasAuthorship W4363671961A5021803359 @default.
- W4363671961 hasAuthorship W4363671961A5044795083 @default.
- W4363671961 hasAuthorship W4363671961A5089509620 @default.
- W4363671961 hasBestOaLocation W43636719611 @default.
- W4363671961 hasConcept C108583219 @default.
- W4363671961 hasConcept C111919701 @default.
- W4363671961 hasConcept C119857082 @default.
- W4363671961 hasConcept C154945302 @default.
- W4363671961 hasConcept C169258074 @default.
- W4363671961 hasConcept C177264268 @default.
- W4363671961 hasConcept C199360897 @default.
- W4363671961 hasConcept C22019652 @default.
- W4363671961 hasConcept C2776760102 @default.
- W4363671961 hasConcept C2984842247 @default.
- W4363671961 hasConcept C41008148 @default.
- W4363671961 hasConcept C50644808 @default.
- W4363671961 hasConcept C519991488 @default.
- W4363671961 hasConcept C98045186 @default.
- W4363671961 hasConceptScore W4363671961C108583219 @default.
- W4363671961 hasConceptScore W4363671961C111919701 @default.
- W4363671961 hasConceptScore W4363671961C119857082 @default.
- W4363671961 hasConceptScore W4363671961C154945302 @default.
- W4363671961 hasConceptScore W4363671961C169258074 @default.
- W4363671961 hasConceptScore W4363671961C177264268 @default.
- W4363671961 hasConceptScore W4363671961C199360897 @default.
- W4363671961 hasConceptScore W4363671961C22019652 @default.
- W4363671961 hasConceptScore W4363671961C2776760102 @default.
- W4363671961 hasConceptScore W4363671961C2984842247 @default.
- W4363671961 hasConceptScore W4363671961C41008148 @default.
- W4363671961 hasConceptScore W4363671961C50644808 @default.
- W4363671961 hasConceptScore W4363671961C519991488 @default.
- W4363671961 hasConceptScore W4363671961C98045186 @default.
- W4363671961 hasLocation W43636719611 @default.
- W4363671961 hasOpenAccess W4363671961 @default.
- W4363671961 hasPrimaryLocation W43636719611 @default.
- W4363671961 hasRelatedWork W2968586400 @default.
- W4363671961 hasRelatedWork W2989932438 @default.
- W4363671961 hasRelatedWork W3099765033 @default.
- W4363671961 hasRelatedWork W3211546796 @default.
- W4363671961 hasRelatedWork W4223564025 @default.
- W4363671961 hasRelatedWork W4281616679 @default.
- W4363671961 hasRelatedWork W4285802257 @default.
- W4363671961 hasRelatedWork W4360619137 @default.
- W4363671961 hasRelatedWork W4361732492 @default.
- W4363671961 hasRelatedWork W4380075502 @default.
- W4363671961 isParatext "false" @default.
- W4363671961 isRetracted "false" @default.
- W4363671961 workType "article" @default.