TNN7: A Custom Macro Suite for Implementing Highly Optimized Designs of Neuromorphic TNNs
This is an incremental improvement for researchers and engineers designing neuromorphic hardware, focusing on optimizing TNN implementations for better efficiency and scalability.
This work tackles the challenge of efficiently implementing Temporal Neural Networks (TNNs) by proposing TNN7, a suite of custom macros that enhance the design framework, resulting in significant reductions in power, delay, area, and energy-delay product by 14%, 16%, 28%, and 45% on average, and enabling implementations like a 40 uW power, 0.05 mm^2 area TNN for time-series clustering and an 18 mW, 24.63 mm^2 TNN achieving 1% error on MNIST.
Temporal Neural Networks (TNNs), inspired from the mammalian neocortex, exhibit energy-efficient online sensory processing capabilities. Recent works have proposed a microarchitecture framework for implementing TNNs and demonstrated competitive performance on vision and time-series applications. Building on these previous works, this work proposes TNN7, a suite of nine highly optimized custom macros developed using a predictive 7nm Process Design Kit (PDK), to enhance the efficiency, modularity and flexibility of the TNN design framework. TNN prototypes for two applications are used for evaluation of TNN7. An unsupervised time-series clustering TNN delivering competitive performance can be implemented within 40 uW power and 0.05 mm^2 area, while a 4-layer TNN that achieves an MNIST error rate of 1% consumes only 18 mW and 24.63 mm^2. On average, the proposed macros reduce power, delay, area, and energy-delay product by 14%, 16%, 28%, and 45%, respectively. Furthermore, employing TNN7 significantly reduces the synthesis runtime of TNN designs (by more than 3x), allowing for highly-scaled TNN implementations to be realized.