Reproduce “Vector Field Neural Networks” 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
Daniel Vieira, Joao Paixao
The vector field neural network (VFNN) transforms data points along a learned Gaussian-based vector field, then classifies them with logistic regression. On an 80%/20% split of Ionosphere, it ranks best among four compared models for accuracy, AUCROC, and MSE. Across five datasets it is reported best or second-best, but comparisons used basic baselines and uneven tuning.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.