Machine Learning based Memory Load Value Predictor for Multimedia Applications
Alain Aoun, Mahmoud Masadeh, Sofiène Tahar · 2024
Approximate computing (AC) has gained traction as an alternative computing method for energy-efficient processing. This paper proposes the exploitation of AC to address the memory wall. The proposed model predicts the load value using machine learning (ML). Subsequently, the ML model is a load value approximator (LVA) where the generated value is accepted as-is. The proposed LVA was tested under various approximate conditions, where 50% to 95% of the load instructions were approximated using multimedia applications. The peak signal-to-noise ratio (PSNR) exceeded 100 dB in several scenarios.