Universal Approximation of Linear Time-Invariant (LTI) Systems through RNNs: Power of Randomness in Reservoir Computing

arXiv:2308.02464v2 Announce Type: replace-cross
Abstract: Recurrent neural networks (RNNs) are known to be universal approximators of dynamic systems under fairly mild and general assumptions. However, RNNs usually suffer from the issues of vanishing and exploding gradients in standard RNN training. Reservoir computing (RC), a special RNN where the recurrent weights are randomized and left untrained, has been introduced to overcome these issues and has demonstrated superior empirical performance especially in scenarios where training samples are extremely limited. On the other …

abstract approximation arxiv assumptions computing cs.lg cs.sy dynamic eess.sp eess.sy general however linear networks neural networks power randomness recurrent neural networks rnn standard systems through training type universal