Deep processing of structured data

10/03/2018
by   Łukasz Maziarka, et al.
0

We construct a general unified framework for learning representation of structured data, i.e. data which cannot be represented as the fixed-length vectors (e.g. sets, graphs, texts or images of varying sizes). The key factor is played by an intermediate network called SAN (Set Aggregating Network), which maps a structured object to a fixed length vector in a high dimensional latent space. Our main theoretical result shows that for sufficiently large dimension of the latent space, SAN is capable of learning a unique representation for every input example. Experiments demonstrate that replacing pooling operation by SAN in convolutional networks leads to better results in classifying images with different sizes. Moreover, its direct application to text and graph data allows to obtain results close to SOTA, by simpler networks with smaller number of parameters than competitive models.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/05/2021

Universal Approximation of Functions on Sets

Modelling functions of sets, or equivalently, permutation-invariant func...
research
08/07/2019

Structuring Autoencoders

In this paper we propose Structuring AutoEncoders (SAE). SAEs are neural...
research
04/01/2022

LASER: LAtent SpacE Rendering for 2D Visual Localization

We present LASER, an image-based Monte Carlo Localization (MCL) framewor...
research
01/25/2019

On the Limitations of Representing Functions on Sets

Recent work on the representation of functions on sets has considered th...
research
07/22/2020

Learning the Latent Space of Robot Dynamics for Cutting Interaction Inference

Utilization of latent space to capture a lower-dimensional representatio...
research
01/03/2021

Learning Neural Networks on SVD Boosted Latent Spaces for Semantic Classification

The availability of large amounts of data and compelling computation pow...
research
11/06/2021

Neural BRDFs: Representation and Operations

Bidirectional reflectance distribution functions (BRDFs) are pervasively...

Please sign up or login with your details

Forgot password? Click here to reset