A Unified Programmable Edge Matrix Processor for Deep Neural Networks and Matrix Algebra

Abstract

Matrix Algebra and Deep Neural Networks represent foundational classes of computational algorithms across multiple emerging applications like Augmented Reality(AR) or Virtual Reality(VR), autonomous navigation (cars, drones, robots), data science, and various artificial intelligence-driven solutions. An accelerator-based architecture can provide performance and energy efficiency supporting fixed functions through customized data paths. However, constrained Edge systems requiring multiple applications and diverse matrix operations to be efficiently supported, cannot afford numerous custom accelerators. In this paper, we present MxCore, a unified architecture that comprises tightly coupled vector and programmable cores sharing data through highly optimized interconnects along with a configurable hardware scheduler managing the co-execution. We submit MxCore as the generalized approach to facilitate the flexible acceleration of multiple Matrix Algebra and Deep-learning applications across a range of sparsity levels. Unified compute resources improve overall resource utilization and performance per unit area. Aggressive and novel microarchitecture techniques along with block-level sparsity support optimize compute and data-reuse to minimize bandwidth and power requirements enabling ultra-low latency applications for low power and cost-sensitive Edge deployments. MxCore requires a small silicon footprint of 0.2068 mm 2 , in a modern 7nm process at 1GHz and achieves (0.15 FP32 and 0.62 INT8) TMAC/ mm 2 , dissipating only 11.66uW of leakage power. At iso-technology and iso-frequency, MxCore provides an energy efficiency of 651.4x, 159.9x, 104.8x, and 124.2x as compared to the 128-core Nvidia’s Maxwell GPU for dense General Matrix Multiply (GEMM), sparse Deep Neural Network (DNN), Cholesky decomposition, and triangular matrix solve respectively.

Modeling and forecasting of stock market processes

“…For this reason, the use of advanced machine learning and artificial intelligence technologies becomes extremely important. With this in mind, the Python library Keras stands out as a powerful tool for building neural networks, in particular multilayer architectures such as LSTM (Long Short-Term Memory) [9, 11], [21,22], [23,24], [25]. LSTMs are one of the most efficient types of recurrent neural networks that are specifically designed for analyzing sequential data, such as time series, which is typical for market data [19,20], [21,22], [23,24], [25,26], [27,28], [29,30].…”

“…With this in mind, the Python library Keras stands out as a powerful tool for building neural networks, in particular multilayer architectures such as LSTM (Long Short-Term Memory) [9, 11], [21,22], [23,24], [25]. LSTMs are one of the most efficient types of recurrent neural networks that are specifically designed for analyzing sequential data, such as time series, which is typical for market data [19,20], [21,22], [23,24], [25,26], [27,28], [29,30]. Therefore, for more efficient analysis, it is necessary to have an LSTM neural network architecture consisting of at least several layers, especially if the amount of information is large.…”