Multiplierless and Sparse Machine Learning based on Margin Propagation Networks
This work addresses the need for ultra-energy-efficient ML processors for IoT and edge computing platforms, offering a novel hardware-software codesign that is incremental in its application to existing classifiers.
The paper tackles the problem of high energy consumption in machine learning hardware by proposing multiplierless and sparse architectures based on margin-propagation networks, which replace matrix-vector multiplications with simple addition and thresholding operations, resulting in comparable performance on benchmark datasets with reduced computational complexity and improved energy efficiency.
The new generation of machine learning processors have evolved from multi-core and parallel architectures that were designed to efficiently implement matrix-vector-multiplications (MVMs). This is because at the fundamental level, neural network and machine learning operations extensively use MVM operations and hardware compilers exploit the inherent parallelism in MVM operations to achieve hardware acceleration on GPUs and FPGAs. However, many IoT and edge computing platforms require embedded ML devices close to the network in order to compensate for communication cost and latency. Hence a natural question to ask is whether MVM operations are even necessary to implement ML algorithms and whether simpler hardware primitives can be used to implement an ultra-energy-efficient ML processor/architecture. In this paper we propose an alternate hardware-software codesign of ML and neural network architectures where instead of using MVM operations and non-linear activation functions, the architecture only uses simple addition and thresholding operations to implement inference and learning. At the core of the proposed approach is margin-propagation (MP) based computation that maps multiplications into additions and additions into a dynamic rectifying-linear-unit (ReLU) operations. This mapping results in significant improvement in computational and hence energy cost. In this paper, we show how the MP network formulation can be applied for designing linear classifiers, shallow multi-layer perceptrons and support vector networks suitable fot IoT platforms and tiny ML applications. We show that these MP based classifiers give comparable results to that of their traditional counterparts for benchmark UCI datasets, with the added advantage of reduction in computational complexity enabling an improvement in energy efficiency.