Reproduce “Design and Analysis of Temporal Representations in 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
Alban Colas-Gattepaille, Alexandre Muzy
DeNN uses learned synaptic delays and spike-arrival times to make event timing shape computation and learning. The authors define temporal functions as those that explicitly manipulate time, unlike weighted activations and binary spike counts. On Google Speech Commands, DeNN reports 98.94% accuracy on 35 classes, versus AST’s 98.11%; this reported comparison is not a matched rerun of every baseline.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.