Landmark Diffusion Maps (L-dMaps): Accelerated manifold learning out-of-sample extension
This work addresses the problem of slow out-of-sample extensions for researchers and practitioners applying diffusion maps to high-volume or high-velocity streaming data, representing an incremental improvement.
The paper tackles the computational bottleneck of out-of-sample extensions in diffusion maps for manifold learning by proposing landmark diffusion maps (L-dMaps), which reduce complexity from O(N) to O(M) with M << N, achieving up to 50-fold speedups and less than 4% errors in reconstruction fidelity on molecular datasets.
Diffusion maps are a nonlinear manifold learning technique based on harmonic analysis of a diffusion process over the data. Out-of-sample extensions with computational complexity $\mathcal{O}(N)$, where $N$ is the number of points comprising the manifold, frustrate applications to online learning applications requiring rapid embedding of high-dimensional data streams. We propose landmark diffusion maps (L-dMaps) to reduce the complexity to $\mathcal{O}(M)$, where $M \ll N$ is the number of landmark points selected using pruned spanning trees or k-medoids. Offering $(N/M)$ speedups in out-of-sample extension, L-dMaps enables the application of diffusion maps to high-volume and/or high-velocity streaming data. We illustrate our approach on three datasets: the Swiss roll, molecular simulations of a C$_{24}$H$_{50}$ polymer chain, and biomolecular simulations of alanine dipeptide. We demonstrate up to 50-fold speedups in out-of-sample extension for the molecular systems with less than 4% errors in manifold reconstruction fidelity relative to calculations over the full dataset.