Evolutionary Neural Architecture Search Supporting Approximate Multipliers
This work addresses the need for automated, energy-efficient neural network design for hardware implementations, representing an incremental improvement in NAS methods by integrating approximate computing.
The paper tackles the problem of designing energy-efficient convolutional neural networks (CNNs) by proposing a multi-objective evolutionary neural architecture search (NAS) method that automatically evolves CNN architectures and selects approximate multipliers to optimize accuracy, network size, and power consumption, achieving competitive results on the CIFAR-10 benchmark.
There is a growing interest in automated neural architecture search (NAS) methods. They are employed to routinely deliver high-quality neural network architectures for various challenging data sets and reduce the designer's effort. The NAS methods utilizing multi-objective evolutionary algorithms are especially useful when the objective is not only to minimize the network error but also to minimize the number of parameters (weights) or power consumption of the inference phase. We propose a multi-objective NAS method based on Cartesian genetic programming for evolving convolutional neural networks (CNN). The method allows approximate operations to be used in CNNs to reduce the power consumption of a target hardware implementation. During the NAS process, a suitable CNN architecture is evolved together with approximate multipliers to deliver the best trade-offs between the accuracy, network size, and power consumption. The most suitable approximate multipliers are automatically selected from a library of approximate multipliers. Evolved CNNs are compared with common human-created CNNs of a similar complexity on the CIFAR-10 benchmark problem.