Analog Optical Inference on Million-Record Mortgage Data

TL;DR

This paper benchmarks an analog optical computer on a large mortgage dataset, analyzing accuracy limitations and potential improvements for machine learning inference.

Contribution

It provides the first large-scale demonstration of analog optical inference on real-world data and identifies key hardware and architectural limitations.

Findings

AOC achieves 94.6% accuracy on mortgage data, close to digital models.

Accuracy gap narrows slightly when increasing optical channels, indicating architectural limits.

Encoding and hardware non-idealities contribute to accuracy loss, guiding future improvements.

Abstract

Analog optical computers promise large efficiency gains for machine learning inference, yet no demonstration has moved beyond small-scale image benchmarks. We benchmark the analog optical computer (AOC) digital twin on mortgage approval classification from 5.84 million U.S. HMDA records and separate three sources of accuracy loss. On the original 19 features, the AOC reaches 94.6% balanced accuracy with 5,126 parameters (1,024 optical), compared with 97.9% for XGBoost; the 3.3 percentage-point gap narrows by only 0.5pp when the optical core is widened from 16 to 48 channels, suggesting an architectural rather than hardware limitation. Restricting all models to a shared 127-bit binary encoding drops every model to 89.4--89.6%, with an encoding cost of 8pp for digital models and 5pp for the AOC. Seven calibrated hardware non-idealities impose no measurable penalty. The three resulting layers of limitation (encoding, architecture, hardware fidelity) locate where accuracy is lost and what to improve next.