C-BRIC JUMP e-Workshop

Abstract: The rapid proliferation of the Internet of Things (IoT) and the dramatic resurgence of artificial intelligence (AI) based application workloads have led to immense interest in performing inference on energy-constrained edge devices. Approximate computing (a design paradigm that trades off a small degradation in application quality for disproportionate energy savings) is a promising technique to enable energy-efficient inference at the edge. This e-Workshop talk introduces the concept of an approximate edge inference system and describes our work in designing a systematic methodology to perform joint approximations between different subsystems in a deep neural network (DNN)-based edge inference system, leading to significant energy benefits compared to approximating individual subsystems in isolation. We use a smart camera system that executes various DNN-based image classification and object detection applications to illustrate how the sensor, memory, compute, and communication subsystems can all be approximated synergistically. We have prototyped such an approximate inference system using an Intel Stratix IV GX-based Terasic TR4-230 FPGA development board. Experimental results obtained using six large DNNs and four compact DNNs running image classification applications demonstrate significant energy savings (≈1.6×–4.7× for large DNNs and ≈1.5×–3.6× for small DNNs) for minimal (<1%) loss in application-level quality. Furthermore, results using four object detection DNNs exhibit energy savings of ≈1.5×–5.2× for similar quality loss. The talk ends by describing our current efforts in extending system-level approximations across multi-device systems in the context of distributed inference.

Bio: Vijay Raghunathan is a Professor in the School of Electrical and Computer Engineering at Purdue University, West Lafayette. He lead the Embedded Systems and IoT Lab (ESL), where their research focuses on hardware and software architectures for embedded systems, wireless sensors for the Internet of Things (IoT), and wearable and implantable electronics, with an emphasis on low power design (at the board-level as well as system-on-chip), micro-scale energy harvesting, emerging memory technologies, and reliable/secure system design.