Journal
FRONTIERS IN NEUROSCIENCE
Volume 15, Issue -, Pages -Publisher
FRONTIERS MEDIA SA
DOI: 10.3389/fnins.2021.629892
Keywords
backpropagation; deep neural networks; weight transport; update locking; edge computing; biologically-plausible learning
Categories
Funding
- National Foundation for Scientific Research (FNRS) of Belgium [1117116F-1117118F]
Ask authors/readers for more resources
The direct random target projection (DRTP) algorithm proposed in this work views one-hot-encoded labels in supervised classification problems as a proxy for error signs, enabling layerwise feedforward training of hidden layers to solve weight transport and update locking issues while reducing computational and memory requirements.
While the backpropagation of error algorithm enables deep neural network training, it implies (i) bidirectional synaptic weight transport and (ii) update locking until the forward and backward passes are completed. Not only do these constraints preclude biological plausibility, but they also hinder the development of low-cost adaptive smart sensors at the edge, as they severely constrain memory accesses and entail buffering overhead. In this work, we show that the one-hot-encoded labels provided in supervised classification problems, denoted as targets, can be viewed as a proxy for the error sign. Therefore, their fixed random projections enable a layerwise feedforward training of the hidden layers, thus solving the weight transport and update locking problems while relaxing the computational and memory requirements. Based on these observations, we propose the direct random target projection (DRTP) algorithm and demonstrate that it provides a tradeoff between accuracy and computational cost that is suitable for adaptive edge computing devices.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available