Institution profile

Austrian Academy of Sciences

Academic institutioneurope · at
Official website
Research library14linked papers
Opportunities0open roles
Selected work

Representative Papers

Training Set Synthesis for Bioacoustic Denoising: A Case Study With Mice

Aug 10, 2026

This study addresses the challenge of denoising bioacoustic recordings corrupted by environmental noise, where clean reference signals are typically unavailable for supervised training. To circumvent the need for real clean data, the authors propose a self-supervised approach that synthesizes training samples containing fundamental frequency and harmonic ridges. They develop a U-Net-based model to predict complex ratio masks and introduce a ridge-guided weighted loss function to better preserve fine-grained vocal structure during denoising. Evaluated on murine ultrasonic vocalizations, the method significantly improves the accuracy of fundamental frequency and harmonic tracking, enhances scale-invariant signal-to-noise ratio, and boosts the generalization performance of downstream classifiers in noisy field conditions.

0 citationsRead paper

Approximation Rates for Metaplectic Neural Networks

Aug 09, 2026

This work addresses the challenge that conventional neural networks struggle to effectively approximate solutions to problems with symplectic structures or quantum dynamics, such as the time-dependent Schrödinger equation. It introduces the metaplectic transform into neural network theory for the first time, constructing a neural dictionary based on this transform and defining a corresponding metaplectic Barron space. The study establishes embedding relations between this space and Sobolev spaces, providing a theoretical foundation for a novel deep network architecture. This architecture leverages finite linear combinations to achieve Monte Carlo approximation of metaplectic Barron functions. Numerical experiments demonstrate that the proposed method significantly outperforms classical physics-informed neural networks in solving the time-dependent Schrödinger equation, thereby validating the expressive power and effectiveness of the introduced dictionary.

0 citationsRead paper

Localizing broadband noise sources using the Loève spectrum and a 2.5D approach

Jun 01, 2026

This study addresses the challenge of accurately localizing uniformly moving broadband random noise sources, a task where conventional methods often fail due to their reliance on signal modification or short-time stationarity assumptions required for Doppler compensation. To overcome these limitations, this work proposes a novel frequency-domain localization approach that integrates Loève spectral theory with a 2.5D acoustic model. For the first time, Loève spectrum theory is applied to the localization of moving broadband sources, establishing a direct mapping between the power spectral density of a moving source and stationary receivers without requiring signal preprocessing or local stationarity assumptions. By incorporating multitaper spectral estimation, the method successfully localizes spectrally flat, stationary broadband sources in simulations involving source velocities up to 100 m/s, thereby surpassing the performance constraints of existing techniques.

0 citationsRead paper
Recent publications

Latest Papers

Training Set Synthesis for Bioacoustic Denoising: A Case Study With Mice

Aug 10, 2026

This study addresses the challenge of denoising bioacoustic recordings corrupted by environmental noise, where clean reference signals are typically unavailable for supervised training. To circumvent the need for real clean data, the authors propose a self-supervised approach that synthesizes training samples containing fundamental frequency and harmonic ridges. They develop a U-Net-based model to predict complex ratio masks and introduce a ridge-guided weighted loss function to better preserve fine-grained vocal structure during denoising. Evaluated on murine ultrasonic vocalizations, the method significantly improves the accuracy of fundamental frequency and harmonic tracking, enhances scale-invariant signal-to-noise ratio, and boosts the generalization performance of downstream classifiers in noisy field conditions.

0 citationsRead paper

Approximation Rates for Metaplectic Neural Networks

Aug 09, 2026

This work addresses the challenge that conventional neural networks struggle to effectively approximate solutions to problems with symplectic structures or quantum dynamics, such as the time-dependent Schrödinger equation. It introduces the metaplectic transform into neural network theory for the first time, constructing a neural dictionary based on this transform and defining a corresponding metaplectic Barron space. The study establishes embedding relations between this space and Sobolev spaces, providing a theoretical foundation for a novel deep network architecture. This architecture leverages finite linear combinations to achieve Monte Carlo approximation of metaplectic Barron functions. Numerical experiments demonstrate that the proposed method significantly outperforms classical physics-informed neural networks in solving the time-dependent Schrödinger equation, thereby validating the expressive power and effectiveness of the introduced dictionary.

0 citationsRead paper

Localizing broadband noise sources using the Loève spectrum and a 2.5D approach

Jun 01, 2026

This study addresses the challenge of accurately localizing uniformly moving broadband random noise sources, a task where conventional methods often fail due to their reliance on signal modification or short-time stationarity assumptions required for Doppler compensation. To overcome these limitations, this work proposes a novel frequency-domain localization approach that integrates Loève spectral theory with a 2.5D acoustic model. For the first time, Loève spectrum theory is applied to the localization of moving broadband sources, establishing a direct mapping between the power spectral density of a moving source and stationary receivers without requiring signal preprocessing or local stationarity assumptions. By incorporating multitaper spectral estimation, the method successfully localizes spectrally flat, stationary broadband sources in simulations involving source velocities up to 100 m/s, thereby surpassing the performance constraints of existing techniques.

0 citationsRead paper