Hybrid neural networks for continual learning inspired by corticohippocampal circuits
Abstract Current artificial systems suffer from catastrophic forgetting during continual learning, a limitation absent in biological systems. Biological mechanisms leverage the dual representation of specific and generalized memories within corticohippocampal circuits to facilitate lifelong learning...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-02-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-025-56405-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571507131285504 |
|---|---|
| author | Qianqian Shi Faqiang Liu Hongyi Li Guangyu Li Luping Shi Rong Zhao |
| author_facet | Qianqian Shi Faqiang Liu Hongyi Li Guangyu Li Luping Shi Rong Zhao |
| author_sort | Qianqian Shi |
| collection | DOAJ |
| description | Abstract Current artificial systems suffer from catastrophic forgetting during continual learning, a limitation absent in biological systems. Biological mechanisms leverage the dual representation of specific and generalized memories within corticohippocampal circuits to facilitate lifelong learning. Inspired by this, we develop a corticohippocampal circuits-based hybrid neural network (CH-HNN) that emulates these dual representations, significantly mitigating catastrophic forgetting in both task-incremental and class-incremental learning scenarios. Our CH-HNNs incorporate artificial neural networks and spiking neural networks, leveraging prior knowledge to facilitate new concept learning through episode inference, and offering insights into the neural functions of both feedforward and feedback loops within corticohippocampal circuits. Crucially, CH-HNN operates as a task-agnostic system without increasing memory demands, demonstrating adaptability and robustness in real-world applications. Coupled with the low power consumption inherent to SNNs, our model represents the potential for energy-efficient, continual learning in dynamic environments. |
| format | Article |
| id | doaj-art-3effc037a29441feae496083878c00a6 |
| institution | Kabale University |
| issn | 2041-1723 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-3effc037a29441feae496083878c00a62025-02-02T12:31:30ZengNature PortfolioNature Communications2041-17232025-02-0116111410.1038/s41467-025-56405-9Hybrid neural networks for continual learning inspired by corticohippocampal circuitsQianqian Shi0Faqiang Liu1Hongyi Li2Guangyu Li3Luping Shi4Rong Zhao5Center for Brain-Inspired Computing Research (CBICR), Tsinghua UniversityCenter for Brain-Inspired Computing Research (CBICR), Tsinghua UniversityCenter for Brain-Inspired Computing Research (CBICR), Tsinghua UniversityCenter for Brain-Inspired Computing Research (CBICR), Tsinghua UniversityCenter for Brain-Inspired Computing Research (CBICR), Tsinghua UniversityCenter for Brain-Inspired Computing Research (CBICR), Tsinghua UniversityAbstract Current artificial systems suffer from catastrophic forgetting during continual learning, a limitation absent in biological systems. Biological mechanisms leverage the dual representation of specific and generalized memories within corticohippocampal circuits to facilitate lifelong learning. Inspired by this, we develop a corticohippocampal circuits-based hybrid neural network (CH-HNN) that emulates these dual representations, significantly mitigating catastrophic forgetting in both task-incremental and class-incremental learning scenarios. Our CH-HNNs incorporate artificial neural networks and spiking neural networks, leveraging prior knowledge to facilitate new concept learning through episode inference, and offering insights into the neural functions of both feedforward and feedback loops within corticohippocampal circuits. Crucially, CH-HNN operates as a task-agnostic system without increasing memory demands, demonstrating adaptability and robustness in real-world applications. Coupled with the low power consumption inherent to SNNs, our model represents the potential for energy-efficient, continual learning in dynamic environments.https://doi.org/10.1038/s41467-025-56405-9 |
| spellingShingle | Qianqian Shi Faqiang Liu Hongyi Li Guangyu Li Luping Shi Rong Zhao Hybrid neural networks for continual learning inspired by corticohippocampal circuits Nature Communications |
| title | Hybrid neural networks for continual learning inspired by corticohippocampal circuits |
| title_full | Hybrid neural networks for continual learning inspired by corticohippocampal circuits |
| title_fullStr | Hybrid neural networks for continual learning inspired by corticohippocampal circuits |
| title_full_unstemmed | Hybrid neural networks for continual learning inspired by corticohippocampal circuits |
| title_short | Hybrid neural networks for continual learning inspired by corticohippocampal circuits |
| title_sort | hybrid neural networks for continual learning inspired by corticohippocampal circuits |
| url | https://doi.org/10.1038/s41467-025-56405-9 |
| work_keys_str_mv | AT qianqianshi hybridneuralnetworksforcontinuallearninginspiredbycorticohippocampalcircuits AT faqiangliu hybridneuralnetworksforcontinuallearninginspiredbycorticohippocampalcircuits AT hongyili hybridneuralnetworksforcontinuallearninginspiredbycorticohippocampalcircuits AT guangyuli hybridneuralnetworksforcontinuallearninginspiredbycorticohippocampalcircuits AT lupingshi hybridneuralnetworksforcontinuallearninginspiredbycorticohippocampalcircuits AT rongzhao hybridneuralnetworksforcontinuallearninginspiredbycorticohippocampalcircuits |