Training of a 2 → 100 → 2 → 5 fully connected ReLU neural net via cross-entropy minimisation. • it starts outputting small embeddings • around epoch 300 learns an identity function • takes 1700 epochs more to unwind the data manifold

Tomorrow, thursday, feb 20th, 11am EDT, Andreas Berentzen @andreasberentzen.bsky.social and Jesse Wonnink will present their Reproducibility study of: Explaining Temporal Graph Models Through an Explorer-Navigator Framework (TMLR and NeurIPS Journal Track). 🪩 zoom link on website