Distributed Adaptive Nearest Neighbor Classifier: Algorithm and Theory
This work addresses classification challenges for large-scale or geographically dispersed datasets, offering an incremental improvement over existing distributed nearest neighbor methods.
The authors tackled the problem of classification with extremely large or distributed data by proposing a distributed adaptive nearest neighbor classifier with a data-driven stochastic tuning parameter and an early stopping rule, achieving nearly optimal convergence rates when sub-sample sizes are large, as demonstrated in simulations and a real-world dataset.
When data is of an extraordinarily large size or physically stored in different locations, the distributed nearest neighbor (NN) classifier is an attractive tool for classification. We propose a novel distributed adaptive NN classifier for which the number of nearest neighbors is a tuning parameter stochastically chosen by a data-driven criterion. An early stopping rule is proposed when searching for the optimal tuning parameter, which not only speeds up the computation but also improves the finite sample performance of the proposed Algorithm. Convergence rate of excess risk of the distributed adaptive NN classifier is investigated under various sub-sample size compositions. In particular, we show that when the sub-sample sizes are sufficiently large, the proposed classifier achieves the nearly optimal convergence rate. Effectiveness of the proposed approach is demonstrated through simulation studies as well as an empirical application to a real-world dataset.