# Automated Circuit Approximation Method Driven by Data Distribution

## TL;DR

This paper introduces a data-driven, automated circuit approximation method using genetic programming and a weighted error metric, optimizing for application-specific accuracy and power efficiency.

## Contribution

It presents a novel approach that translates application-level error metrics into component-level errors using WMED, enabling efficient circuit approximation tailored to data distribution.

## Key findings

- Effective approximation of MAC units for neural networks
- Improved trade-offs between accuracy and power consumption
- Validated on synthetic benchmarks and real classifiers

## Abstract

We propose an application-tailored data-driven fully automated method for functional approximation of combinational circuits. We demonstrate how an application-level error metric such as the classification accuracy can be translated to a component-level error metric needed for an efficient and fast search in the space of approximate low-level components that are used in the application. This is possible by employing a weighted mean error distance (WMED) metric for steering the circuit approximation process which is conducted by means of genetic programming. WMED introduces a set of weights (calculated from the data distribution measured on a selected signal in a given application) determining the importance of each input vector for the approximation process. The method is evaluated using synthetic benchmarks and application-specific approximate MAC (multiply-and-accumulate) units that are designed to provide the best trade-offs between the classification accuracy and power consumption of two image classifiers based on neural networks.