Training Neural Networks for Execution on Approximate Hardware
This work addresses the need for efficient training methods for approximate hardware, particularly for battery-operated devices, but it appears incremental as it builds on existing approximate computing methods.
The paper tackles the problem of training neural networks for execution on approximate hardware, which lacks specialized training methods, and demonstrates that training can be sped up by up to 18X.
Approximate computing methods have shown great potential for deep learning. Due to the reduced hardware costs, these methods are especially suitable for inference tasks on battery-operated devices that are constrained by their power budget. However, approximate computing hasn't reached its full potential due to the lack of work on training methods. In this work, we discuss training methods for approximate hardware. We demonstrate how training needs to be specialized for approximate hardware, and propose methods to speed up the training process by up to 18X.