Reproduce “Truncated Variational EM for Semi-Supervised Neural Simpletrons” 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
Dennis Forster, Jörg Lücke
The authors apply truncated variational EM to a generative handwritten-digit network, retaining each input’s 15 highest-scoring templates from 10,000 and zeroing the rest. On MNIST with 10 labeled examples total, t-NeSi reports 7.22% ± 0.53% test error, versus 10.91% ± 0.86% without truncation. However, the truncation setting was tuned using 10 labels per class.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.