Interpretable Super-Resolution via a Learned Time-Series Representation

06/13/2020
by   Randall Balestriero, et al.
0

We develop an interpretable and learnable Wigner-Ville distribution that produces a super-resolved quadratic signal representation for time-series analysis. Our approach has two main hallmarks. First, it interpolates between known time-frequency representations (TFRs) in that it can reach super-resolution with increased time and frequency resolution beyond what the Heisenberg uncertainty principle prescribes and thus beyond commonly employed TFRs, Second, it is interpretable thanks to an explicit low-dimensional and physical parameterization of the Wigner-Ville distribution. We demonstrate that our approach is able to learn highly adapted TFRs and is ready and able to tackle various large-scale classification tasks, where we reach state-of-the-art performance compared to baseline and learned TFRs.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 13

page 14

page 15

07/24/2018

Classification with Joint Time-Frequency Scattering

In time series classification, signals are typically mapped into some in...
06/01/2019

Super-resolution of Time-series Labels for Bootstrapped Event Detection

Solving real-world problems, particularly with deep learning, relies on ...
10/30/2021

Learning Continuous Representation of Audio for Arbitrary Scale Super Resolution

Audio super resolution aims to predict the missing high resolution compo...
04/24/2020

Mining self-similarity: Label super-resolution with epitomic representations

We show that simple patch-based models, such as epitomes, can have super...
11/05/2018

Blind Two-Dimensional Super-Resolution and Its Performance Guarantee

Super-resolution techniques are concerned with extracting fine-scale dat...
05/20/2018

DLBI: Deep learning guided Bayesian inference for structure reconstruction of super-resolution fluorescence microscopy

Super-resolution fluorescence microscopy, with a resolution beyond the d...
07/15/2020

timeXplain – A Framework for Explaining the Predictions of Time Series Classifiers

Modern time series classifiers display impressive predictive capabilitie...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.