Reproduce “Toward stochastic neural computing” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yang Qi, Zhichao Zhu, Yiming Wei, Lu Cao, Zhigang Wang, Jie Zhang, Wenlian Lu, Jianfeng Feng
The moment neural network (MNN) trains spiking networks by propagating each population’s mean firing rate and spike-count covariance, then reconstructs the network without post-training tuning. On MNIST, the simulated network reaches a 0.980 probability of correct prediction after 6.6 ms with 300 hidden-layer spikes, versus a 0.9845 unlimited-readout limit. It retains only second-order statistics and was demonstrated on a feedforward network.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.