Neural Advances in Processing Nonlinear Dynamic Signals
5 papers5 directions
Neural Advances in Processing Nonlinear Dynamic Signals is a research venue focused on the intersection of neural network architectures and nonlinear signal processing. Active in 2020, the venue published work primarily centered on hierarchical temporal integration within deep recurrent networks. Its contributions explore the mathematical foundations and practical applications of processing complex, time-varying data through specialized machine learning models.
The research published at this venue covers a range of technical topics, including linear reservoir computing and the application of four-dimensional hypercomplex algebras in adaptive signal processing. Several papers address audio-specific challenges, such as the separation of drum and bass components from monaural tracks and the development of low-complexity, linear-in-the-parameters filters for distorted speech signals. Additionally, the venue features work on optimizing neural components, such as learning activation functions from data using cubic spline interpolation.
Key contributors to this venue include researchers such as Aurelio Uncini, Danilo Comminiello, Michele Scarpiniti, and Simone Scardapane, alongside Alessio Micheli and Claudio Gallicchio. By bridging the gap between traditional signal processing techniques and modern deep learning, the venue provided a platform for discussing efficient temporal representations and advanced filtering methods. Its output reflects a specific interest in the computational efficiency and mathematical rigor of nonlinear dynamic systems within the machine learning community.
Papers by Year
| 2020 | 5 |
| 2017 | 1 |
| 2016 | 1 |
Top Research Directions
Research directions most represented by papers from this venue.
Top Authors
| Author | Papers |
|---|---|
| Danilo Comminiello | 4 |
| Aurelio Uncini | 3 |
| Simone Scardapane | 3 |
| Alessio Micheli | 1 |
| Claudio Gallicchio | 1 |