Reproduce “MinMaxLTTB: Leveraging MinMax-Preselection to Scale LTTB” 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

Jeroen Van Der Donckt, Jonas Van Der Donckt, Michael Rademaker, Sofie Van Hoecke

MinMaxLTTB accelerates downsampling for large time-series charts by selecting local minima and maxima, then applying LTTB to that shortlist. It retains the original endpoints but still scans the input, so both methods scale linearly. Benchmarks reported up to 10× faster single-core and 30× faster parallel execution than the tested LTTB implementation; image metrics were often close, not proof of perceptual equivalence.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.