Visible to the public Short and Squeezed: Accelerating the Computation of Antisparse Representations with Safe Squeezing

TitleShort and Squeezed: Accelerating the Computation of Antisparse Representations with Safe Squeezing
Publication TypeConference Paper
Year of Publication2020
AuthorsElvira, Clément, Herzet, Cédric
Conference NameICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
KeywordsAcceleration, antisparse coding, Collaboration, convex optimization, Geometry, Human Behavior, human factors, Metrics, numerical simulation, Optimization, policy-based governance, pubcrawl, resilience, Resiliency, Safe Coding, safe squeezing, scaled projected gradient, Signal processing, speech processing, Transforms
AbstractAntisparse coding aims at spreading the information uniformly over representation coefficients and can be expressed as the solution of an ℓ∞-norm regularized problem. In this paper, we propose a new methodology, coined "safe squeezing", accelerating the computation of antisparse representations. The idea consists in identifying saturated entries of the solution via simple tests and compacting their contribution to achieve some form of dimensionality reduction. Numerical experiments show that the proposed approach leads to significant computational gain.
DOI10.1109/ICASSP40776.2020.9053156
Citation Keyelvira_short_2020