Deep Learning with Sets and Point Clouds
This work addresses the challenge of processing unordered sets in machine learning, offering a novel method for tasks like outlier detection and semi-supervised learning, though it is incremental in advancing set-based architectures.
The paper tackles the problem of handling set-structured data in deep learning by introducing a permutation equivariant layer, achieving linear-time complexity and demonstrating applications in point cloud classification and MNIST-digit summation with invariant outputs.
We introduce a simple permutation equivariant layer for deep learning with set structure.This type of layer, obtained by parameter-sharing, has a simple implementation and linear-time complexity in the size of each set. We use deep permutation-invariant networks to perform point-could classification and MNIST-digit summation, where in both cases the output is invariant to permutations of the input. In a semi-supervised setting, where the goal is make predictions for each instance within a set, we demonstrate the usefulness of this type of layer in set-outlier detection as well as semi-supervised learning with clustering side-information.