ePoster · Machine Learning
Efficient learning of deep non-negative matrix factorisation networks
Mahbod Nouri, David Rotermund, Alberto García Ortiz, Klaus Pawelzik · Bernstein Conference 2024
Networks composed of Non-Negative Matrix Factorization (NNMF) [1] modules can serve as abstractions of real neural networks. In particular, NNMF networks can be extended easily to perform computations based on stochastic spikes [2,3]. However, due to training via Autograd approaches following exact gradients to train deep networks of NNMF modules currently requires prohibitively large amounts of memory and is very slow. Here we present an approximative backpropagation (BP) method for optimizing deep NNMF networks. NNMF layers have latent variables that are updated iteratively towards a fixed