Jun, 2020
一种实用的稀疏逼近算法用于实时循环学习
A Practical Sparse Approximation for Real Time Recurrent Learning
Jacob Menick, Erich Elsen, Utku Evci, Simon Osindero, Karen Simonyan...
TL;DR介绍了 Sparse n-step Approximation (SnAp) 来优化 Real Time Recurrent Learning (RTRL) 的影响矩阵,在保持计算成本低的同时提高了网络的学习速度,特别是在稀疏网络时,n=2 时的 SnAp 能够优于 backpropagation in terms of learning speed when updates are done online。