Interpretable Locally Adaptive Nearest Neighbors
This work addresses the need for interpretable and adaptive metrics in machine learning, offering a domain-specific improvement over global metric methods.
The paper tackled the problem of learning locally adaptive metrics for k-nearest neighbors algorithms, resulting in improved performance and natural interpretability, as demonstrated on synthetic and real-world benchmark datasets.
When training automated systems, it has been shown to be beneficial to adapt the representation of data by learning a problem-specific metric. This metric is global. We extend this idea and, for the widely used family of k nearest neighbors algorithms, develop a method that allows learning locally adaptive metrics. These local metrics not only improve performance but are naturally interpretable. To demonstrate important aspects of how our approach works, we conduct a number of experiments on synthetic data sets, and we show its usefulness on real-world benchmark data sets.