A Fixed Size Storage O ( n 3 ) Time Complexity Learning Algorithm for Fully Recurrent Continually Running Networks
1992; The MIT Press; Volume: 4; Issue: 2 Linguagem: Inglês
10.1162/neco.1992.4.2.243
ISSN1530-888X
Autores Tópico(s)Energy Efficient Wireless Sensor Networks
ResumoThe real-time recurrent learning (RTRL) algorithm (Robinson and Fallside 1987; Williams and Zipser 1989) requires O(n 4 ) computations per time step, where n is the number of noninput units. I describe a method suited for on-line learning that computes exactly the same gradient and requires fixed-size storage of the same order but has an average time complexity per time step of O(n 3 ).
Referência(s)