FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3033597970> ?p ?o ?g. }

• W3033597970 abstract "Recurrent neural networks (RNNs) with continuous-time hidden states are a natural fit for modeling irregularly-sampled time series. These models, however, face difficulties when the input data possess long-term dependencies. We prove that similar to standard RNNs, the underlying reason for this issue is the vanishing or exploding of the gradient during training. This phenomenon is expressed by the ordinary differential equation (ODE) representation of the hidden state, regardless of the ODE solver's choice. We provide a solution by designing a new algorithm based on the long short-term memory (LSTM) that separates its memory from its time-continuous state. This way, we encode a continuous-time dynamical flow within the RNN, allowing it to respond to inputs arriving at arbitrary time-lags while ensuring a constant error propagation through the memory path. We call these RNN models ODE-LSTMs. We experimentally show that ODE-LSTMs outperform advanced RNN-based counterparts on non-uniformly sampled data with long-term dependencies. All code and data is available at this https URL." @default.
• W3033597970 created "2020-06-12" @default.
• W3033597970 creator A5025684214 @default.
• W3033597970 creator A5036588940 @default.
• W3033597970 date "2020-06-08" @default.
• W3033597970 modified "2023-09-27" @default.
• W3033597970 title "Learning Long-Term Dependencies in Irregularly-Sampled Time Series" @default.
• W3033597970 cites W1498436455 @default.
• W3033597970 cites W1587799944 @default.
• W3033597970 cites W1689711448 @default.
• W3033597970 cites W1815076433 @default.
• W3033597970 cites W1924770834 @default.
• W3033597970 cites W1971129545 @default.
• W3033597970 cites W2012964785 @default.
• W3033597970 cites W2018074000 @default.
• W3033597970 cites W2058580716 @default.
• W3033597970 cites W2064675550 @default.
• W3033597970 cites W2107878631 @default.
• W3033597970 cites W2112796928 @default.
• W3033597970 cites W2116261113 @default.
• W3033597970 cites W2131774270 @default.
• W3033597970 cites W2144144709 @default.
• W3033597970 cites W2150355110 @default.
• W3033597970 cites W2158782408 @default.
• W3033597970 cites W2295663533 @default.
• W3033597970 cites W2547418827 @default.
• W3033597970 cites W2551482042 @default.
• W3033597970 cites W2736601468 @default.
• W3033597970 cites W2761228892 @default.
• W3033597970 cites W2811040116 @default.
• W3033597970 cites W2885195348 @default.
• W3033597970 cites W2930930439 @default.
• W3033597970 cites W2948216679 @default.
• W3033597970 cites W2962751418 @default.
• W3033597970 cites W2962817261 @default.
• W3033597970 cites W2963094322 @default.
• W3033597970 cites W2963255584 @default.
• W3033597970 cites W2963673225 @default.
• W3033597970 cites W2963755523 @default.
• W3033597970 cites W2964010366 @default.
• W3033597970 cites W2964244673 @default.
• W3033597970 cites W2968352263 @default.
• W3033597970 cites W2970181183 @default.
• W3033597970 cites W2970241199 @default.
• W3033597970 cites W2970898247 @default.
• W3033597970 cites W2970971581 @default.
• W3033597970 cites W2971278153 @default.
• W3033597970 cites W2971306341 @default.
• W3033597970 cites W2971343405 @default.
• W3033597970 cites W2994942528 @default.
• W3033597970 cites W2995001833 @default.
• W3033597970 cites W2996160951 @default.
• W3033597970 cites W3034500979 @default.
• W3033597970 cites W3090132620 @default.
• W3033597970 cites W3096541186 @default.
• W3033597970 cites W3096986660 @default.
• W3033597970 hasPublicationYear "2020" @default.
• W3033597970 type Work @default.
• W3033597970 sameAs 3033597970 @default.
• W3033597970 citedByCount "0" @default.
• W3033597970 crossrefType "posted-content" @default.
• W3033597970 hasAuthorship W3033597970A5025684214 @default.
• W3033597970 hasAuthorship W3033597970A5036588940 @default.
• W3033597970 hasConcept C11413529 @default.
• W3033597970 hasConcept C119857082 @default.
• W3033597970 hasConcept C121332964 @default.
• W3033597970 hasConcept C134306372 @default.
• W3033597970 hasConcept C143724316 @default.
• W3033597970 hasConcept C147168706 @default.
• W3033597970 hasConcept C151406439 @default.
• W3033597970 hasConcept C151730666 @default.
• W3033597970 hasConcept C154945302 @default.
• W3033597970 hasConcept C177264268 @default.
• W3033597970 hasConcept C17744445 @default.
• W3033597970 hasConcept C199360897 @default.
• W3033597970 hasConcept C199539241 @default.
• W3033597970 hasConcept C2776359362 @default.
• W3033597970 hasConcept C2776760102 @default.
• W3033597970 hasConcept C2777735758 @default.
• W3033597970 hasConcept C2778112365 @default.
• W3033597970 hasConcept C2778770139 @default.
• W3033597970 hasConcept C28826006 @default.
• W3033597970 hasConcept C33923547 @default.
• W3033597970 hasConcept C34862557 @default.
• W3033597970 hasConcept C41008148 @default.
• W3033597970 hasConcept C50644808 @default.
• W3033597970 hasConcept C51544822 @default.
• W3033597970 hasConcept C54355233 @default.
• W3033597970 hasConcept C61797465 @default.
• W3033597970 hasConcept C62520636 @default.
• W3033597970 hasConcept C78045399 @default.
• W3033597970 hasConcept C86803240 @default.
• W3033597970 hasConcept C94625758 @default.
• W3033597970 hasConceptScore W3033597970C11413529 @default.
• W3033597970 hasConceptScore W3033597970C119857082 @default.
• W3033597970 hasConceptScore W3033597970C121332964 @default.
• W3033597970 hasConceptScore W3033597970C134306372 @default.
• W3033597970 hasConceptScore W3033597970C143724316 @default.
• W3033597970 hasConceptScore W3033597970C147168706 @default.
• W3033597970 hasConceptScore W3033597970C151406439 @default.

• next