Control Variate Approximation for DNN Accelerators
TimeTuesday, December 7th3:50pm - 4:10pm PST
Event Type
Research Manuscript
Virtual Programs
Presented In-Person
Approximate Computing for AI/ML
DescriptionIn this work, we introduce a control variate approximation technique for low error approximate Deep Neural Network (DNN) accelerators. The control variate technique is used in Monte Carlo methods to achieve variance reduction. Our approach significantly decreases the induced error due to approximate multiplications in DNN inference, without requiring time-exhaustive retraining compared to state-of-the-art.
Leveraging our control variate method, we use highly approximated multipliers to generate power-optimized DNN accelerators. Our experimental evaluation on six DNNs, for Cifar-10 and Cifar-100 datasets, demonstrates that our control variate approximation achieves 24% power reduction for a merely 0.16% accuracy loss.