Space Expansion of Feature Selection for Designing more Accurate Error Predictors

TL;DR

This paper introduces a scheduling-aware feature selection method for error predictors in approximate computing, enhancing prediction accuracy by utilizing intermediate hardware results and system constraints.

Contribution

It proposes a novel feature selection approach that leverages hardware intermediate results and system parameters to improve error prediction accuracy in approximate computing.

Findings

Significant improvement in prediction accuracy over prior methods.

Flexible prediction timing for higher accuracy.

Effective trade-off between energy, latency, and prediction quality.

Abstract

Approximate computing is being considered as a promising design paradigm to overcome the energy and performance challenges in computationally demanding applications. If the case where the accuracy can be configured, the quality level versus energy efficiency or delay also may be traded-off. For this technique to be used, one needs to make sure a satisfactory user experience. This requires employing error predictors to detect unacceptable approximation errors. In this work, we propose a scheduling-aware feature selection method which leverages the intermediate results of the hardware accelerator to improve the prediction accuracy. Additionally, it configures the error predictors according to the energy consumption and latency of the system. The approach enjoys the flexibility of the prediction time for a higher accuracy. The results on various benchmarks demonstrate significant improvements in the prediction accuracy compared to the prior works which used only the accelerator inputs for the prediction.