Interpretable Super-Resolution via a Learned Time-Series Representation
This addresses the need for interpretable, high-resolution signal analysis in time-series applications, though it appears incremental as it builds on existing time-frequency representations.
The paper tackles the problem of achieving super-resolution in time-frequency representations for time-series analysis by developing an interpretable learned Wigner-Ville distribution, which reaches state-of-the-art performance on large-scale classification tasks.
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.