Learning Generative Models of Similarity Matrices
This work addresses the challenge of improving clustering algorithms for researchers and practitioners in machine learning, offering a novel framework that can lead to more robust methods, though it appears incremental as it builds upon existing spectral clustering concepts.
The authors tackled the problem of data clustering by proposing a probabilistic generative model for similarity matrices, which provides a framework for developing new algorithms and extended models. They demonstrated that their new generative clustering models perform well on a broader class of problems where other methods like spectral clustering often fail, showing excellent performance in experiments on various point data sets.
We describe a probabilistic (generative) view of affinity matrices along with inference algorithms for a subclass of problems associated with data clustering. This probabilistic view is helpful in understanding different models and algorithms that are based on affinity functions OF the data. IN particular, we show how(greedy) inference FOR a specific probabilistic model IS equivalent TO the spectral clustering algorithm.It also provides a framework FOR developing new algorithms AND extended models. AS one CASE, we present new generative data clustering models that allow us TO infer the underlying distance measure suitable for the clustering problem at hand. These models seem to perform well in a larger class of problems for which other clustering algorithms (including spectral clustering) usually fail. Experimental evaluation was performed in a variety point data sets, showing excellent performance.