Instabilities in Convnets for Raw Audio - Archive ouverte HAL Access content directly
Journal Articles IEEE Signal Processing Letters Year : 2024

Instabilities in Convnets for Raw Audio

Abstract

What makes waveform-based deep learning so hard? Despite numerous attempts at training convolutional neural networks (convnets) for filterbank design, they often fail to outperform hand-crafted baselines. These baselines are linear time-invariant systems: as such, they can be approximated by convnets with wide receptive fields. Yet, in practice, gradient-based optimization leads to suboptimal approximations. In our article, we approach this phenomenon from the perspective of initialization. We present a theory of large deviations for the energy response of FIR filterbanks with random Gaussian weights. We find that deviations worsen for large filters and locally periodic input signals, which are both typical for audio signal processing applications. Numerical simulations align with our theory and suggest that the condition number of a convolutional layer follows a logarithmic scaling law between the number and length of the filters, which is reminiscent of discrete wavelet bases.
Fichier principal
Vignette du fichier
haider2024spl.pdf (548.58 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04528116 , version 1 (04-04-2024)

Licence

Identifiers

Cite

Daniel Haider, Vincent Lostanlen, Martin Ehler, Peter Balazs. Instabilities in Convnets for Raw Audio. IEEE Signal Processing Letters, 2024, 31, pp.1084-1088. ⟨10.1109/LSP.2024.3386492⟩. ⟨hal-04528116⟩
18 View
29 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More