Provably noise-robust, regularised $k$-means clustering
This addresses the problem of robust clustering for data analysts in noisy environments, offering incremental improvements over prior methods.
The paper tackles clustering in noisy data by proposing a regularized k-means method that is provably robust to unstructured points, improving separation requirements from δ > 2√2 + ε to δ > 2 + ε under the stochastic ball model and showing better performance than k-means++ on MNIST with noise.
We consider the problem of clustering in the presence of noise. That is, when on top of cluster structure, the data also contains a subset of \emph{unstructured} points. Our goal is to detect the clusters despite the presence of many unstructured points. Any algorithm that achieves this goal is noise-robust. We consider a regularisation method which converts any center-based clustering objective into a noise-robust one. We focus on the $k$-means objective and we prove that the regularised version of $k$-means is NP-Hard even for $k=1$. We consider two algorithms based on the convex (sdp and lp) relaxation of the regularised objective and prove robustness guarantees for both. The sdp and lp relaxation of the standard (non-regularised) $k$-means objective has been previously studied by [ABC+15]. Under the stochastic ball model of the data they show that the sdp-based algorithm recovers the underlying structure as long as the balls are separated by $δ> 2\sqrt{2} + ε$. We improve upon this result in two ways. First, we show recovery even for $δ> 2 + ε$. Second, our regularised algorithm recovers the balls even in the presence of noise so long as the number of noisy points is not too large. We complement our theoretical analysis with simulations and analyse the effect of various parameters like regularization constant, noise-level etc. on the performance of our algorithm. In the presence of noise, our algorithm performs better than $k$-means++ on MNIST.