NAX: Neural Architecture and Memristive Xbar based Accelerator Co-design
TimeWednesday, July 13th10:30am - 10:53am PDT
Location3002, Level 3
Event Type
Research Manuscript
In-memory and Near-memory Computing
DescriptionNeural Architecture Search (NAS) has provided the ability to design efficient deep neural network (DNN) catered towards different hardware like GPUs, CPUs etc. However, integrating NAS with Memristive Crossbar Array (MCA) based In-Memory Computing (IMC) hardware remains an open problem. Co-optimization of both network and hardware parameters presents a challenging search space comprising of different kernel sizes mapped to varying crossbar sizes. To that effect, we propose NAX – an efficient NAS engine that co-designs neural network and IMC based hardware architecture. Finally, We perform experiments on CIFAR-10 and Tiny ImageNet dataset to show the effectiveness for the framework.