Efficient Design of Hardware-Enabled Recurrent Neural Networks
share
In this work, we propose a new approach towards the efficient design of reservoir computing hardware. First, we adapt the reservoir input mask to the structure of the data via linear autoencoders. We therefore incorporate the advantages of dimensionality reduction and dimensionality expansion achieved by conventional and efficient linear algebra procedures of principal component analysis. Second, we employ evolutionary-inspired genetic algorithm techniques resulting in a highly efficient optimization of reservoir dynamics. We illustrate the method on the so-called single-node reservoir computing architecture, especially suitable for implementation in ultrahigh-speed hardware. The combination of both methods and the resulting reduction of time required for performance optimization of a hardware system establish a strategy towards machine learning hardware capable of self-adaption to optimally solve specific problems. We confirm the validity of those principles building RC hardware based on a field-programmable gate array.
READ FULL TEXT