Running different embedding algorithms on the spiral data set (top), in which points are sampled from a unit-speed parameterized Archimedean spiral. Different choices of the neighborhood size, , may produce qualitatively different results, depending on the algorithm. Running those same algorithms using the IAN kernel (right) typically gives a reasonable result. Refer to the main text for details.
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.