# At-Scale Evaluation of Weight Clustering to Enable Energy-Efficient Object Detection

## TL;DR

This paper evaluates weight clustering at scale for YOLOv3 object detection on real driving videos, demonstrating significant reductions in bandwidth and energy consumption with minimal accuracy loss, informing accelerator design.

## Contribution

It provides a comprehensive at-scale assessment of weight clustering for YOLOv3 in real-world driving scenarios, highlighting its effectiveness in energy and bandwidth savings.

## Key findings

- Weight clustering reduces bandwidth to 30-40% of original.
- Energy consumption decreases by approximately 55%.
- Layer-wise clustering with 5-8 bits maintains near-original accuracy.

## Abstract

Accelerators implementing Deep Neural Networks for image-based object detection operate on large volumes of data due to fetching images and neural network parameters, especially if they need to process video streams, hence with high power dissipation and bandwidth requirements to fetch all those data. While some solutions exist to mitigate power and bandwidth demands for data fetching, they are often assessed in the context of limited evaluations with a scale much smaller than that of the target application, which challenges finding the best tradeoff in practice. This paper sets up the infrastructure to assess at-scale a key power and bandwidth optimization - weight clustering - for You Only Look Once v3 (YOLOv3), a neural network-based object detection system, using videos of real driving conditions. Our assessment shows that accelerators such as systolic arrays with an Output Stationary architecture turn out to be a highly effective solution combined with weight clustering. In particular, applying weight clustering independently per neural network layer, and using between 32 (5-bit) and 256 (8-bit) weights allows achieving an accuracy close to that of the original YOLOv3 weights (32-bit weights). Such bit-count reduction of the weights allows shaving bandwidth requirements down to 30%-40% of the original requirements, and reduces energy consumption down to 45%. This is based on the fact that (i) energy due to multiply-and-accumulate operations is much smaller than DRAM data fetching, and (ii) designing accelerators appropriately may make that most of the data fetched corresponds to neural network weights, where clustering can be applied. Overall, our at-scale assessment provides key results to architect camera-based object detection accelerators by putting together a real-life application (YOLOv3), and real driving videos, in a unified setup so that trends observed are reliable.