paper-with-me

홈 › Papers

Semi-Periodic Activation for Time Series Classification

2024-12-13 · José Gilberto Barbosa de Medeiros Júnior, Andre Guarnier de Mitri, Diego Furtado Silva

This paper investigates the lack of research on activation functions for neural network models in time series tasks. It highlights the need to identify essential properties of these activations to improve their effectiveness in specific domains. To this end, the study comprehensively analyzes properties, such as bounded, monotonic, nonlinearity, and periodicity, for activation in time series neural networks. We propose a new activation that maximizes the coverage of these properties, called LeakySineLU. We empirically evaluate the LeakySineLU against commonly used activations in the literature using 112 benchmark datasets for time series classification, obtaining the best average ranking in all comparative scenarios.

📄 PDF Abstract BibTeX arXiv:2412.09889

Code (1)

jose-gilberto/leakysinelu 공식 구현 pytorch

Tasks

ClassificationTime SeriesTime Series Classification

Similar Papers 제목 키워드 기반

Enhancing Representation Learning for Periodic Time Series with Floss: A Frequency Domain Regularization Approach

2023-08-02 · Chunwei Yang, Xiaoxu Chen, Lijun Sun, Hongyu Yang 외

Time series analysis is a fundamental task in various application domains, and deep learning approaches have demonstrated remarkable performance in this area. However, many real-world time series data exhibit significant…

Anomaly DetectionDeep LearningRepresentation LearningTime Series+2

Neural Decomposition of Time-Series Data for Effective Generalization

2017-05-25 · Luke B. Godfrey, Michael S. Gashler

We present a neural network technique for the analysis and extrapolation of time-series data called Neural Decomposition (ND). Units with a sinusoidal activation function are used to perform a Fourier-like decomposition …

Time SeriesTime Series AnalysisTime Series Forecasting

From Taylor Series to Fourier Synthesis: The Periodic Linear Unit

2025-08-02 · Shiko Kudo arxiv

The dominant paradigm in modern neural networks relies on simple, monotonically-increasing activation functions like ReLU. While effective, this paradigm necessitates large, massively-parameterized models to approximate …

How neural networks learn to classify chaotic time series

2023-06-04 · Alessandro Corbetta, Thomas Geert de Jong

Neural networks are increasingly employed to model, analyze and control non-linear dynamical systems ranging from physics to biology. Owing to their universal approximation capabilities, they regularly outperform state-o…

Time Series

Periodic Activation Functions Induce Stationarity

2021-10-26 · NeurIPS 2021 12 · Lassi Meronen, Martin Trapp, Arno Solin

Neural network models are known to reinforce hidden data biases, making them unreliable and difficult to interpret. We seek to build models that `know what they do not know' by introducing inductive biases in the functio…

SensitivityTranslation