Reproduce “DxNAT - Deep Neural Networks for Explaining Non-Recurring Traffic Congestion” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Fangzhou sun, Abhishek Dubey, Jules White
DxNAT turns road speeds into map-shaped images and uses a convolutional neural network to classify traffic images as non-recurring or recurring congestion, with crossover augmentation. In Nashville football-game validation, it reports 98.73% accuracy versus 84.06% for a random forest on the same split. The validation used one game day and one non-game day, limiting the scope of this result.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.