Analog Optical Inference on Million-Record Mortgage Data
For the analog optical computing community, this provides the first large-scale, real-world benchmark on million-record data, revealing that accuracy gaps are due to architectural limitations rather than hardware non-idealities.
The paper benchmarks an analog optical computer (AOC) digital twin on mortgage approval classification using 5.84 million HMDA records, achieving 94.6% balanced accuracy with 5,126 parameters, compared to 97.9% for XGBoost. It identifies three layers of accuracy loss: encoding (5pp for AOC, 8pp for digital), architecture (3.3pp gap), and hardware fidelity (no measurable penalty).
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.