TeMPO: Efficient Time-Multiplexed Dynamic Photonic Tensor Core for Edge AI with Compact Slow-Light Electro-Optic Modulator

TL;DR

TeMPO is a novel photonic tensor core that leverages domain-specific device and architecture customization to achieve high-performance, energy-efficient AI inference on edge devices, bridging the gap with electronic accelerators.

Contribution

The paper introduces TeMPO, a time-multiplexed photonic accelerator with customized slow-light modulators and a multi-core architecture, significantly improving edge AI performance and efficiency.

Findings

Achieves 368.6 TOPS peak performance

Attains 22.3 TOPS/W energy efficiency

Presents a foundry-compatible photonic device design

Abstract

Electronic-photonic computing systems offer immense potential in energy-efficient artificial intelligence (AI) acceleration tasks due to the superior computing speed and efficiency of optics, especially for real-time, low-energy deep neural network (DNN) inference tasks on resource-restricted edge platforms. However, current optical neural accelerators based on foundry-available devices and conventional system architecture still encounter a performance gap compared to highly customized electronic counterparts. To bridge the performance gap due to lack of domain specialization, we present a time-multiplexed dynamic photonic tensor accelerator, dubbed TeMPO, with cross-layer device/circuit/architecture customization. At the device level, we present foundry-compatible, customized photonic devices, including a slow-light electro-optic modulator with experimental demonstration, optical splitters, and phase shifters that significantly reduce the footprint and power in input encoding and dot-product calculation. At the circuit level, partial products are hierarchically accumulated via parallel photocurrent aggregation, lightweight capacitive temporal integration, and sequential digital summation, considerably relieving the analog-to-digital conversion bottleneck. We also employ a multi-tile, multi-core architecture to maximize hardware sharing for higher efficiency. Across diverse edge AI workloads, TeMPO delivers digital-comparable task accuracy with superior quantization/noise tolerance. We achieve a 368.6 TOPS peak performance, 22.3 TOPS/W energy efficiency, and 1.2 TOPS/mm$^2$ compute density, pushing the Pareto frontier in edge AI hardware. This work signifies the power of cross-layer co-design and domain-specific customization, paving the way for future electronic-photonic accelerators with even greater performance and efficiency.