A brain-inspired neural network-based echo state network (BINN-ESN) that uses a modified continuous coupled neural network (MCCNN) as the neural model, which is inspired by the mammalian visual cortex for time-series prediction.
Abstract
Summary Time-series prediction based on historical data is essential in numerous scientific fields, such as weather prediction and financial markets analysis. However, obtaining strong predictive accuracy together with high computational efficiency remains challenging. To address this challenge, we propose a brain-inspired neural network-based echo state network (BINN-ESN). It uses a modified continuous coupled neural network (MCCNN) as the neural model, which is inspired by the mammalian visual cortex. Our results indicate a system-dependent trade-off: While deep learning baselines generally achieve stronger single-step accuracy, BINN-ESN shows stronger long-term predictive stability on most evaluated systems and requires substantially less total computation than the GPU-accelerated long short-term memory (LSTM) baseline in long-term experiments. Our code is available at https://github.com/Jizhao-Liu/code-for-BINN-ESN.
A novel architecture that combines a graph neural network block with a Long Short-Term Memoryencoder-decoder block for end-to-end feature learning and forecasting in multivariate time series forecasting is proposed, resulting in accurate and reliable predictions.
Ali Sarabi, Arash Sarabi, G. Runger· IISE Annual Conference &...· 0 citations
Spiking neural networks (SNNs), as core models of brain-inspired computing, simulate the computational mechanisms of biological neural systems. However, existing models lack the division of labor and collaboration mechanisms found in biological fast and slow neural populations, making it difficult for single-timescale...
Jun-Xiu Liu, Zhao-Xin Zhang, Qiang Fu et al.· Neural Networks· 0 citations
Multivariate time series data constitutes the fundamental basis for decision making across real-world scenarios such as financial trading, intensive care, energy dispatch and the Internet of Things. Traditional statistical methods and early deep learning architectures including recurrent neural networks and convolution...
De-Liang Zhang, Bo He, Yu-Xin Gao· Journal of Computing and Ele...· 0 citations
Reliable rainfall forecasting across both short-term (hourly to daily) and long-term (weekly to seasonal) horizons remains a central challenge for climate prediction systems, hydrological planning, disaster risk reduction, and agricultural decision-making. Conventional deep learning architectures such as convolutional...
Sahul Hameed S· MSRDG International Journal...· 0 citations
Frequency Selective Neural Network is introduced, pioneering a foundation architecture guaranteeing physical interpretability without sacrificing expressive power of deep learning, and converges directly on physically meaningful frequency bands, providing a highly scalable, interpretable paradigm for robust pattern rec...
Data-driven machine-learning weather models now rival, and in some respects surpass, operational numerical weather prediction (NWP), yet no single model dominates across variables, pressure levels, lead times, or regions, and the marginal returns from developing ever-larger individual models are diminishing. We present...
