This repository contains the official implementation of "Adaptive Dendritic Plasticity in Brain-inspired Dynamic Neural Networks for Enhanced Multi-timescale Feature Extraction" (Under Review at IEEE TNNLS).
🚧 This repository is under construction. The code will be made publicly available soon.
We propose a novel brain-inspired dynamic neural network with two key innovations:
- DH-LIM: A Dendritic Heterogeneity Leaky Integrate Modulate neuron model that replaces traditional binary spike activation with a continuous modulation mechanism
- ADP: An Adaptive Dendritic Plasticity mechanism that dynamically adjusts dendritic decay factors based on input signal frequency characteristics
- Enhanced multi-timescale feature extraction capabilities
- State-of-the-art performance on EEG and ECG temporal sequence recognition tasks
- Efficient and biologically plausible neural dynamics
- Comprehensive evaluation on DEAP, ECG, and Alzheimer's disease datasets
If you find this work useful for your research, please consider citing:
@article{mao2025adaptive,
title={Adaptive Dendritic Plasticity in Brain-inspired Dynamic Neural Networks for Enhanced Multi-timescale Feature Extraction},
author={Mao, Jiayi},
journal={IEEE Transactions on Neural Networks and Learning Systems},
note={Under Review},
year={2025}
}For any questions or discussions, please feel free to contact:
- Jiayi Mao (maojy23@mails.tsinghua.edu.cn)
This project is licensed under the MIT License - see the LICENSE file for details.
We thank all reviewers for their valuable feedback. This work was supported by the Center for Brain Inspired Computing Research, Tsinghua University.