How Powerful Are Randomly Initialized Pointcloud Set Functions?

Abstract

We study random embeddings produced by untrained neural set functions, and show that they are powerful representations which well capture the input features for downstream tasks such as classification, and are often linearly separable. We obtain surprising results that show that random set functions can often obtain close to or even better accuracy than fully trained models. We investigate factors that affect the representative power of such embeddings quantitatively and qualitatively.

arXiv comment6 pages, 2 figures, NeurIPS 2019 Sets & Partitions Workshop

References in corpus (5)

• Instance Normalization: The Missing Ingredient for Fast Stylization
• Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling
• Learning Representations and Generative Models for 3D Point Clouds
• No Training Required: Exploring Random Encoders for Sentence Classification
• A Powerful Generative Model Using Random Weights for the Deep Image Representation