Learning Neural Models for End-to-End Clustering

TL;DR

This paper introduces a neural network architecture capable of performing end-to-end probabilistic clustering on various data types, trained in a supervised manner to generalize across different datasets and grouping criteria.

Contribution

It presents a novel end-to-end neural clustering model that estimates distributions over cluster counts and assignments, differing from existing methods by enabling learnable, fully end-to-end clustering.

Findings

Promising results on image and speech datasets

Able to generalize to different data and grouping criteria

Distinct from deep metric learning and semi-supervised clustering

Abstract

We propose a novel end-to-end neural network architecture that, once trained, directly outputs a probabilistic clustering of a batch of input examples in one pass. It estimates a distribution over the number of clusters $k$, and for each $1 \leq k \leq k_\mathrm{max}$, a distribution over the individual cluster assignment for each data point. The network is trained in advance in a supervised fashion on separate data to learn grouping by any perceptual similarity criterion based on pairwise labels (same/different group). It can then be applied to different data containing different groups. We demonstrate promising performance on high-dimensional data like images (COIL-100) and speech (TIMIT). We call this ``learning to cluster'' and show its conceptual difference to deep metric learning, semi-supervise clustering and other related approaches while having the advantage of performing learnable clustering fully end-to-end.