SoftHebb: Bayesian Inference in Unsupervised Hebbian Soft Winner-Take-All Networks. (arXiv:2107.05747v3 [cs.LG] UPDATED)

Hebbian plasticity in winner-take-all (WTA) networks is highly attractive for
neuromorphic on-chip learning, owing to its efficient, local, unsupervised, and
on-line nature. Moreover, its biological plausibility may help overcome
important limitations of artificial algorithms, such as their susceptibility to
adversarial attacks and long training time. However, Hebbian WTA learning has
found little use in machine learning (ML), likely because it has been missing
an optimization theory compatible with deep learning (DL). Here we show
rigorously that WTA networks constructed by …

arxiv bayesian bayesian inference inference lg networks unsupervised