Matches in SemOpenAlex for { <https://semopenalex.org/work/W3119096943> ?p ?o ?g. }
- W3119096943 abstract "The success of Convolutional Neural Networks (CNNs) in computer vision is mainly driven by their strong inductive bias, which is strong enough to allow CNNs to solve vision-related tasks with random weights, meaning without learning. Similarly, Long Short-Term Memory (LSTM) has a strong inductive bias towards storing information over time. However, many real-world systems are governed by conservation laws, which lead to the redistribution of particular quantities -- e.g. in physical and economical systems. Our novel Mass-Conserving LSTM (MC-LSTM) adheres to these conservation laws by extending the inductive bias of LSTM to model the redistribution of those stored quantities. MC-LSTMs set a new state-of-the-art for neural arithmetic units at learning arithmetic operations, such as addition tasks, which have a strong conservation law, as the sum is constant over time. Further, MC-LSTM is applied to traffic forecasting, modelling a pendulum, and a large benchmark dataset in hydrology, where it sets a new state-of-the-art for predicting peak flows. In the hydrology example, we show that MC-LSTM states correlate with real-world processes and are therefore interpretable." @default.
- W3119096943 created "2021-01-18" @default.
- W3119096943 creator A5005882134 @default.
- W3119096943 creator A5033763438 @default.
- W3119096943 creator A5037001244 @default.
- W3119096943 creator A5053148274 @default.
- W3119096943 creator A5053159516 @default.
- W3119096943 creator A5056778863 @default.
- W3119096943 creator A5069068123 @default.
- W3119096943 creator A5079632405 @default.
- W3119096943 date "2021-01-13" @default.
- W3119096943 modified "2023-09-27" @default.
- W3119096943 title "MC-LSTM: Mass-Conserving LSTM" @default.
- W3119096943 cites W1505551852 @default.
- W3119096943 cites W1538131130 @default.
- W3119096943 cites W1787224781 @default.
- W3119096943 cites W1836465849 @default.
- W3119096943 cites W1979769287 @default.
- W3119096943 cites W1987440467 @default.
- W3119096943 cites W1990211566 @default.
- W3119096943 cites W2001317094 @default.
- W3119096943 cites W2004286722 @default.
- W3119096943 cites W2033904036 @default.
- W3119096943 cites W2041579970 @default.
- W3119096943 cites W2064675550 @default.
- W3119096943 cites W2069291384 @default.
- W3119096943 cites W2076063813 @default.
- W3119096943 cites W2088174986 @default.
- W3119096943 cites W2088704524 @default.
- W3119096943 cites W2089217417 @default.
- W3119096943 cites W2100641602 @default.
- W3119096943 cites W2100649405 @default.
- W3119096943 cites W2101926813 @default.
- W3119096943 cites W2112796928 @default.
- W3119096943 cites W2127512094 @default.
- W3119096943 cites W2130123495 @default.
- W3119096943 cites W2136848157 @default.
- W3119096943 cites W2138763184 @default.
- W3119096943 cites W2147107577 @default.
- W3119096943 cites W2151307655 @default.
- W3119096943 cites W2341905372 @default.
- W3119096943 cites W2342730350 @default.
- W3119096943 cites W2345859161 @default.
- W3119096943 cites W2547718140 @default.
- W3119096943 cites W2573587735 @default.
- W3119096943 cites W2603766970 @default.
- W3119096943 cites W2608096034 @default.
- W3119096943 cites W2618236535 @default.
- W3119096943 cites W2749795597 @default.
- W3119096943 cites W2781888692 @default.
- W3119096943 cites W2791827404 @default.
- W3119096943 cites W2792491689 @default.
- W3119096943 cites W2798517054 @default.
- W3119096943 cites W2800819102 @default.
- W3119096943 cites W2884775584 @default.
- W3119096943 cites W2885716143 @default.
- W3119096943 cites W2887020936 @default.
- W3119096943 cites W2899283552 @default.
- W3119096943 cites W2912985636 @default.
- W3119096943 cites W2914767245 @default.
- W3119096943 cites W2919115771 @default.
- W3119096943 cites W2921283045 @default.
- W3119096943 cites W2949888546 @default.
- W3119096943 cites W2950193666 @default.
- W3119096943 cites W2950635152 @default.
- W3119096943 cites W2962716426 @default.
- W3119096943 cites W2962899986 @default.
- W3119096943 cites W2963042606 @default.
- W3119096943 cites W2963090522 @default.
- W3119096943 cites W2963172539 @default.
- W3119096943 cites W2963454111 @default.
- W3119096943 cites W2963504252 @default.
- W3119096943 cites W2963836885 @default.
- W3119096943 cites W2964121744 @default.
- W3119096943 cites W2964153729 @default.
- W3119096943 cites W2964278142 @default.
- W3119096943 cites W2970866842 @default.
- W3119096943 cites W2970971581 @default.
- W3119096943 cites W2971162686 @default.
- W3119096943 cites W2971940059 @default.
- W3119096943 cites W2976793329 @default.
- W3119096943 cites W2982585804 @default.
- W3119096943 cites W2989857225 @default.
- W3119096943 cites W2991103574 @default.
- W3119096943 cites W2992035660 @default.
- W3119096943 cites W2995149074 @default.
- W3119096943 cites W2995154758 @default.
- W3119096943 cites W2995894011 @default.
- W3119096943 cites W2996451395 @default.
- W3119096943 cites W3012294986 @default.
- W3119096943 cites W3017050719 @default.
- W3119096943 cites W3024402844 @default.
- W3119096943 cites W3024702011 @default.
- W3119096943 cites W3034944009 @default.
- W3119096943 cites W3101634859 @default.
- W3119096943 cites W3102090196 @default.
- W3119096943 cites W3115065063 @default.
- W3119096943 cites W3157286395 @default.
- W3119096943 cites W36434594 @default.
- W3119096943 cites W648786980 @default.