Weight-Agnostic Neural Networks

Re:Weight-Agnostic Neural Networks by s_Hogg on Thursday June 13, 2019 @07:28AM (#20173690)

Is there meant to be a direct link between this sort of thing and the randomly-initialised approach to Echo State Networks?

[ Parent | on HN ]

Re:Weight-Agnostic Neural Networks by s_Hogg on Thursday June 13, 2019 @07:30AM (#20173705)

apologies, reading further they refer to:

> a randomly-initialized LSTM [12] with a learned linear output layer can predict time series where traditional RNNs fail

[ Parent | on HN ]

Re:Weight-Agnostic Neural Networks by hardmaru on Thursday June 13, 2019 @07:54AM (#20173859)

Yeah, though I think we could have referenced a few more works in the reservoir computing area. Will keep that in mind for the next draft.

[ Parent | on HN ]

Re:Weight-Agnostic Neural Networks by theSage on Thursday June 13, 2019 @08:16AM (#20174010)

just starting to read this. It reminds me of meiosis networks + stochastic delta rule where the weights were drawn from a random distribution (normal I think).

[ Parent | on HN ]

Re:Weight-Agnostic Neural Networks by hardmaru on Thursday June 13, 2019 @11:51PM (#20180465)

Added a few cites to these works, thanks.

[ Parent | on HN ]