Matches in SemOpenAlex for { <https://semopenalex.org/work/W2991310588> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2991310588 abstract "Sequential and temporal data arise in many fields of research, such as quantitative finance, medicine, or computer vision. A novel approach for sequential learning, called the signature method and rooted in rough path theory, is considered. Its basic principle is to represent multidimensional paths by a graded feature set of their iterated integrals, called the signature. This approach relies critically on an embedding principle, which consists in representing discretely sampled data as paths, i.e., functions from $[0,1]$ to $mathbb{R}^d$. After a survey of machine learning methodologies for signatures, the influence of embeddings on prediction accuracy is investigated with an in-depth study of three recent and challenging datasets. It is shown that a specific embedding, called lead-lag, is systematically the strongest performer across all datasets and algorithms considered. Moreover, an empirical study reveals that computing signatures over the whole path domain does not lead to a loss of local information. It is concluded that, with a good embedding, combining signatures with other simple algorithms achieves results competitive with state-of-the-art, domain-specific approaches." @default.
- W2991310588 created "2019-12-05" @default.
- W2991310588 creator A5076921655 @default.
- W2991310588 date "2019-11-29" @default.
- W2991310588 modified "2023-09-25" @default.
- W2991310588 title "Embedding and learning with signatures" @default.
- W2991310588 cites W1544274373 @default.
- W2991310588 cites W1583788335 @default.
- W2991310588 cites W1593543504 @default.
- W2991310588 cites W1678356000 @default.
- W2991310588 cites W1792076819 @default.
- W2991310588 cites W2004095444 @default.
- W2991310588 cites W2038484192 @default.
- W2991310588 cites W2101234009 @default.
- W2991310588 cites W2119818729 @default.
- W2991310588 cites W2131099969 @default.
- W2991310588 cites W2132940048 @default.
- W2991310588 cites W2139794777 @default.
- W2991310588 cites W2143246460 @default.
- W2991310588 cites W2160511393 @default.
- W2991310588 cites W2297500149 @default.
- W2991310588 cites W2410815474 @default.
- W2991310588 cites W2555077524 @default.
- W2991310588 cites W2559200333 @default.
- W2991310588 cites W2612241623 @default.
- W2991310588 cites W2735590100 @default.
- W2991310588 cites W2768236879 @default.
- W2991310588 cites W2786632635 @default.
- W2991310588 cites W2787731280 @default.
- W2991310588 cites W2788385411 @default.
- W2991310588 cites W2894729295 @default.
- W2991310588 cites W2898186212 @default.
- W2991310588 cites W2898407550 @default.
- W2991310588 cites W2911964244 @default.
- W2991310588 cites W2947626232 @default.
- W2991310588 cites W2949390547 @default.
- W2991310588 cites W2962874504 @default.
- W2991310588 cites W2963445321 @default.
- W2991310588 cites W2963611658 @default.
- W2991310588 cites W2994713923 @default.
- W2991310588 cites W3102326534 @default.
- W2991310588 cites W3102476541 @default.
- W2991310588 cites W3106415951 @default.
- W2991310588 cites W3137136951 @default.
- W2991310588 cites W3146166473 @default.
- W2991310588 cites W3146797348 @default.
- W2991310588 cites W337289788 @default.
- W2991310588 cites W617501864 @default.
- W2991310588 hasPublicationYear "2019" @default.
- W2991310588 type Work @default.
- W2991310588 sameAs 2991310588 @default.
- W2991310588 citedByCount "12" @default.
- W2991310588 countsByYear W29913105882020 @default.
- W2991310588 countsByYear W29913105882021 @default.
- W2991310588 countsByYear W29913105882022 @default.
- W2991310588 crossrefType "posted-content" @default.
- W2991310588 hasAuthorship W2991310588A5076921655 @default.
- W2991310588 hasBestOaLocation W29913105881 @default.
- W2991310588 hasConcept C154945302 @default.
- W2991310588 hasConcept C41008148 @default.
- W2991310588 hasConcept C41608201 @default.
- W2991310588 hasConceptScore W2991310588C154945302 @default.
- W2991310588 hasConceptScore W2991310588C41008148 @default.
- W2991310588 hasConceptScore W2991310588C41608201 @default.
- W2991310588 hasLocation W29913105881 @default.
- W2991310588 hasLocation W29913105882 @default.
- W2991310588 hasLocation W29913105883 @default.
- W2991310588 hasOpenAccess W2991310588 @default.
- W2991310588 hasPrimaryLocation W29913105881 @default.
- W2991310588 hasRelatedWork W1503053695 @default.
- W2991310588 hasRelatedWork W2541017027 @default.
- W2991310588 hasRelatedWork W2562338087 @default.
- W2991310588 hasRelatedWork W2781888261 @default.
- W2991310588 hasRelatedWork W2806860970 @default.
- W2991310588 hasRelatedWork W2892722436 @default.
- W2991310588 hasRelatedWork W3091793561 @default.
- W2991310588 hasRelatedWork W3096054746 @default.
- W2991310588 hasRelatedWork W4285211716 @default.
- W2991310588 hasRelatedWork W4285224442 @default.
- W2991310588 isParatext "false" @default.
- W2991310588 isRetracted "false" @default.
- W2991310588 magId "2991310588" @default.
- W2991310588 workType "article" @default.