from Hacker News

A solution to the learning dilemma for recurrent networks of spiking neurons

by aethertap on 1/29/20, 11:47 AM with 1 comments

  • by marmaduke on 2/4/20, 1:13 PM

    I just saw Mass present this paper at HBP summit and came to post it here. It sounds fairly interesting since it can perform gradients without reverse mode differentiation.