Notebooks developed in Mathematica for my Ph.D. thesis and other resources
Github Code for "Noise-Generating and Imaging Mechanism Inspired Implicit Regularization Learning Network for Low Dose CT Reconstrution" (IEEE TMI 2024)
MPGD: Chaotic Regularization and Heavy-Tailed Limits for Deterministic Gradient Descent
Independent PyTorch reproduction of ICR/ECR attribution-robustness training (Mehrpanah et al., ICML 2026) on ResNet18/CIFAR-10, includes a diagnostic that traces a non-reproducing ICR result to the learning-rate sweep never reaching the edge-of-stability regime the mechanism depends on.
Research Project on NMF
A research-driven analysis of dynamic ticket pricing, modeling distributions with scaled Beta estimates derived from limited statistics (min, max, mean, median). The approach enriches Random Forest classification by incorporating shape parameters (α, β) and leveraging constant-value features for implicit regularization. Based on SeatGeek data.
Production-ready framework for training robust computer vision models. Features multi-GPU support, EMA tracking, label smoothing, and comprehensive robustness evaluation across 4 noise types. Includes scalable TF.Data pipeline, automated testing, Docker support, and CLI tools. Install: pip install robust-vision
To associate your repository with the implicit-regularization topic, visit your repo's landing page and select "manage topics."