Investigating the Resilience Source of Classification Systems for Approximate Computing Techniques

Abstract

During the last decade, classification systems (CSs) received significant research attention, with new learning algorithms achieving high accuracy in various applications. However, their resource-intensive nature, in terms of hardware and computation time, poses new design challenges. CSs exhibit inherent error resilience, due to redundancy of training sets, and self-healing properties, making them suitable for Approximate Computing (AxC). AxC enables efficient computation by using reduced precision or approximate values, leading to energy, time, and silicon area savings. Exploiting AxC involves estimating the introduced error for each approximate variant found during a Design-Space Exploration (DSE). This estimation has to be both rapid and meaningful, considering a substantial number of test samples, which are utterly conflicting demands. In this article, we investigate on sources of error resiliency of CSs, and we propose a technique to haste the DSE that reduces the computational time for error estimation by systematically reducing the test set. In particular, we cherry-pick samples that are likely to be more sensitive to approximation and perform accuracy-loss estimation just by exploiting such a sample subset. In order to demonstrate its efficacy, we integrate our technique into two different approaches for generating approximate CSs, showing an average speed-up up to $\approx$18.