TCD-NPE: A Re-configurable and Efficient Neural Processing Engine, Powered by Novel Temporal-Carry-deferring MACs

TL;DR

This paper introduces TCD-NPE, a reconfigurable neural processing engine utilizing novel Temporal-Carry-deferring MACs, achieving significant improvements in energy efficiency and speed for neural network processing.

Contribution

The paper presents the design of TCD-MAC and TCD-NPE, a reconfigurable engine with a novel scheduler that reduces computational rounds and enhances performance.

Findings

TCD-NPE outperforms conventional neural processors in energy efficiency.

TCD-NPE achieves faster execution times.

The proposed scheduler optimizes processing sequences for minimal rounds.

Abstract

In this paper, we first propose the design of Temporal-Carry-deferring MAC (TCD-MAC) and illustrate how our proposed solution can gain significant energy and performance benefit when utilized to process a stream of input data. We then propose using the TCD-MAC to build a reconfigurable, high speed, and low power Neural Processing Engine (TCD-NPE). We, further, propose a novel scheduler that lists the sequence of needed processing events to process an MLP model in the least number of computational rounds in our proposed TCD-NPE. We illustrate that our proposed TCD-NPE significantly outperform similar neural processing solutions that use conventional MACs in terms of both energy consumption and execution time.