Reproduce “DNN-GDITD: Out-of-distribution detection via Deep Neural Network based Gaussian Descriptor for Imbalanced Tabular Data” 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
Priyanka Chudasama, Anil Surisetty, Aakarsh Malhotra, Alok Singh
DNN-GDITD combines imbalanced tabular classification with out-of-distribution (OOD) detection by modeling each known class as a Gaussian-like region in a DNN feature space. It assigns the best-scoring class when any score is nonnegative and rejects samples when all scores are negative. On Gas Sensor at MDSR 0.10, OOD-class AUPR is 96.28%, versus 92.64% for Deep-MCDD, in the tested setup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.