# Bit-Identical Medical Deep Learning via Structured Orthogonal Initialization

arXiv: https://arxiv.org/abs/2603.28040 · HTML: https://mlsanity.com/paper/2603.28040/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 37 of its 54 references that are in the map.
- First posted on arXiv 30 Mar 2026.

## Builds on (most cited first, 30 of 37)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [SGDR: Stochastic Gradient Descent with Warm Restarts](https://mlsanity.com/paper/1608.03983/) (2016)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Conformer: Convolution-augmented Transformer for Speech Recognition](https://mlsanity.com/paper/2005.08100/) (2020)
- [Exact solutions to the nonlinear dynamics of learning in deep linear neural networks](https://mlsanity.com/paper/1312.6120/) (2013)
- [Prevalence of Neural Collapse during the terminal phase of deep learning training](https://mlsanity.com/paper/2008.08186/) (2020)
- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [Improving Reproducibility in Machine Learning Research (A Report from the NeurIPS 2019 Reproducibility Program)](https://mlsanity.com/paper/2003.12206/) (2020)
- [A Geometric Analysis of Neural Collapse with Unconstrained Features](https://mlsanity.com/paper/2105.02375/) (2021)
- [Accounting for Variance in Machine Learning Benchmarks](https://mlsanity.com/paper/2103.03098/) (2021)
- [Fixup Initialization: Residual Learning Without Normalization](https://mlsanity.com/paper/1901.09321/) (2019)
- [fastai: A Layered API for Deep Learning](https://mlsanity.com/paper/2002.04688/) (2020)
- [Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networks](https://mlsanity.com/paper/1806.05393/) (2018)
- [Automatic diagnosis of the 12-lead ECG using a deep neural network](https://mlsanity.com/paper/1904.01949/) (2019)
- [Deep Learning for ECG Analysis: Benchmarks and Insights from PTB-XL](https://mlsanity.com/paper/2004.13701/) (2020)
- [Resurrecting the sigmoid in deep learning through dynamical isometry: theory and practice](https://mlsanity.com/paper/1711.04735/) (2017)
- [All you need is a good init](https://mlsanity.com/paper/1511.06422/) (2015)
- [Inducing Neural Collapse in Imbalanced Learning: Do We Really Need a Learnable Classifier at the End of Deep Neural Network?](https://mlsanity.com/paper/2203.09081/) (2022)
- [Self-supervised representation learning from 12-lead ECG data](https://mlsanity.com/paper/2103.12676/) (2021)
- [Provable Benefit of Orthogonal Initialization in Optimizing Deep Linear Networks](https://mlsanity.com/paper/2001.05992/) (2020)
- [Randomness In Neural Network Training: Characterizing The Impact of Tooling](https://mlsanity.com/paper/2106.11872/) (2021)
- [Reproducibility in Machine Learning-based Research: Overview, Barriers and Drivers](https://mlsanity.com/paper/2406.14325/) (2024)
- [ZerO Initialization: Initializing Neural Networks with only Zeros and Ones](https://mlsanity.com/paper/2110.12661/) (2021)
- [An Electrocardiogram Foundation Model Built on over 10 Million Recordings with External Evaluation across Multiple Domains](https://mlsanity.com/paper/2410.04133/) (2024)
- [Towards Training Reproducible Deep Learning Models](https://mlsanity.com/paper/2202.02326/) (2022)
- [On the Variance of Neural Network Training with respect to Test Sets and Distributions](https://mlsanity.com/paper/2304.01910/) (2023)
- [Enhancing accuracy of deep learning algorithms by training with low-discrepancy sequences](https://mlsanity.com/paper/2005.12564/) (2020)

## Abstract (arXiv)

Deep learning training is non-deterministic: identical code with different random seeds produces models that agree on aggregate metrics but disagree on individual predictions, with per-class AUC swings exceeding 20 percentage points on rare clinical classes. We present a framework for verified bit-identical training that eliminates three sources of randomness: weight initialization (via structured orthogonal basis functions), batch ordering (via golden ratio scheduling), and non-deterministic GPU operations (via architecture selection and custom autograd). The pipeline produces MD5-verified identical trained weights across independent runs. On PTB-XL ECG rhythm classification, structured initialization significantly exceeds Kaiming across two architectures (n=20; Conformer p = 0.016, Baseline p < 0.001), reducing aggregate variance by 2-3x and reducing per-class variability on rare rhythms by up to 7.5x (TRIGU range: 4.1pp vs 30.9pp under Kaiming, independently confirmed by 3-fold CV). A four-basis comparison at n=20 shows all structured orthogonal bases produce equivalent performance (Friedman p=0.48), establishing that the contribution is deterministic structured initialization itself, not any particular basis function. Cross-domain validation on seven MedMNIST benchmarks (n=20, all p > 0.14) confirms no performance penalty on standard tasks; per-class analysis on imbalanced tasks (ChestMNIST, RetinaMNIST) shows the same variance reduction on rare classes observed in ECG. Cross-dataset evaluation on three external ECG databases confirms zero-shot generalization (>0.93 AFIB AUC).