Reproduce “PLEIADES: Building Temporal Kernels with Orthogonal Polynomials” 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
Yan Ru Pei, Olivier Coenen
PLEIADES represents each temporal kernel as learned coefficients over fixed Jacobi orthogonal polynomials, enabling its continuous kernel to be resampled for different event-bin sizes. On DVS128 gesture recognition, a model trained with 10 ms bins was resampled to 5 ms and 20 ms bins at a fixed 100 ms window without fine-tuning; its accuracy–latency curves remained largely unchanged in that experiment.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.