journal article·2026-01

Fiber Transistors as a Hardware Surrogate Gradient for Backpropagation in Spiking Neural Networks

Weichu Chen, Hao Jiang, Chengyang Du, Yueheng Zhong, Xiangyu Wang, Xiang Li, Zhu Chen, Qicheng Liang, Fengqiang Sun, Yuwen Zhu, Jian‐Gang Chen, Liang‐Wen Feng, Hongzhi Wang, Meifang Zhu, Hengda Sun, Gang Wang

Advanced Materials· Volume 38· Issue 6
Publisher page

Abstract

Abstract Efficient training of spiking neural networks (SNNs) in flexible neuromorphic hardware remains a major challenge due to the non‐differentiable nature of spiking activations and the limited availability of trainable, energy‐efficient device platforms. Here, a tunable textile‐based vertical organic electrochemical transistor (TT‐vOECT) enables implementation of surrogate gradients computation for backpropagation in SNNs. Using a solvent interdiffusion solidification spinning strategy, Plateau–Rayleigh instabilities are overcome to fabricate coaxial trilayer fibers with well‐defined heterojunctions between PEDOT:PSS and BBL. The resulting TT‐vOECT exhibits nonlinear transfer characteristics that closely approximate the Sigmoid derivative. A conditionally activated backpropagation (CAB) mechanism is further proposed, in which synaptic updates are gated by both surrogate gradient magnitude and input spike, is implemented using reconfigurable logic arrays based on dual TT‐vOECTs. This framework enables sparse, event‐driven weight updates with reduced computational overhead. Integrated into a convolutional SNN, the device‐driven CAB strategy achieves high‐accuracy classification of electroencephalogram signals for multiclass neurological disorder diagnosis, comparable to software baselines, while reducing computational redundancy by ≈20%. The results establish a scalable and flexible organic hardware platform for trainable neuromorphic systems with biologically inspired learning dynamics.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Advanced Memory and Neural Computing · Neural Networks and Reservoir Computing · Advanced Sensor and Energy Harvesting Materials

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

State Key Laboratory of Advanced Fiber Materials College of Materials Science and Engineering Donghua University Shanghai 201620 China

Shanghai Key Laboratory of Multidimensional Information Processing School of Communication & Electronic Engineering East China Normal University Shanghai 200241 China

Key Laboratory of Green Chemistry & Technology Ministry of Education College of Chemistry Sichuan University Chengdu 610065 China

Henan Academy of Sciences Zhengzhou 450046 China

Yuyue Home Textile Co., Ltd. Shandong 256623 China

Funding

National Key Research and Development Program of China · Science and Technology Commission of Shanghai Municipality · National Natural Science Foundation of China · Fundamental Research Funds for the Central Universities

References (54)

Show source-provided references
1. 10.1038/s41467-024-51110-510.1038/s41467-024-51110-5 ↗
2. 10.1038/s42256-021-00388-x10.1038/s42256-021-00388-x ↗
3. 10.1038/s41928-022-00913-910.1038/s41928-022-00913-9 ↗
4. 10.1038/s41928-019-0350-y10.1038/s41928-019-0350-y ↗
5. 10.1038/s41467-022-28483-610.1038/s41467-022-28483-6 ↗
6. 10.1002/adma.20241574310.1002/adma.202415743 ↗
7. 10.1038/s41586-025-08742-410.1038/s41586-025-08742-4 ↗
8. 10.1038/s41467-024-53827-910.1038/s41467-024-53827-9 ↗
9. 10.1021/acs.chemrev.2c0072010.1021/acs.chemrev.2c00720 ↗
10. 10.1002/adma.20241607310.1002/adma.202416073 ↗
11. 10.1038/s41928-023-01020-z10.1038/s41928-023-01020-z ↗
12. 10.1038/s41928-023-01012-z10.1038/s41928-023-01012-z ↗
13. 10.1126/science.275.5297.21310.1126/science.275.5297.213 ↗
14. 10.1016/j.neuron.2019.05.04110.1016/j.neuron.2019.05.041 ↗
15. 10.1016/j.neunet.2018.12.00210.1016/j.neunet.2018.12.002 ↗
16. 10.1109/MSP.2019.293159510.1109/MSP.2019.2931595 ↗
17. D.Huh T. J.Sejnowski presented atProceedings of the 32nd International Conference on Neural Information Processing Systems 2018 p:1440.
18. 10.1109/JPROC.2023.330808810.1109/JPROC.2023.3308088 ↗
19. 10.1038/s41563-022-01450-810.1038/s41563-022-01450-8 ↗
20. 10.1038/s41467-021-22528-y10.1038/s41467-021-22528-y ↗
21. 10.1038/s41467-024-50496-610.1038/s41467-024-50496-6 ↗
22. 10.1093/nsr/nwaf33110.1093/nsr/nwaf331 ↗
23. 10.1038/s41467-022-29773-910.1038/s41467-022-29773-9 ↗
24. 10.1038/ncomms840910.1038/ncomms8409 ↗
25. 10.1073/pnas.200612211710.1073/pnas.2006122117 ↗
26. 10.1002/adma.20200345310.1002/adma.202003453 ↗
27. 10.1038/s41586-024-07343-x10.1038/s41586-024-07343-x ↗
28. 10.1038/s41586-025-09344-w10.1038/s41586-025-09344-w ↗
29. 10.1126/science.adk375510.1126/science.adk3755 ↗
30. Gupta A. K. · IEEE Trans. Circuit. Systems Ii‐Express Briefs · 2005 · 52 · 10
31. 10.1073/pnas.210919411910.1073/pnas.2109194119 ↗
32. N.Perez‐Nieves D. F. M.Goodman presented at35th Annual Conference on Neural Information Processing Systems (NeurIPS) 2021 p:11795.
33. 10.1021/acs.chemmater.3c0223710.1021/acs.chemmater.3c02237 ↗
34. 10.1002/adma.20110479510.1002/adma.201104795 ↗
35. 10.1021/am404113n10.1021/am404113n ↗
36. 10.1126/sciadv.adi353610.1126/sciadv.adi3536 ↗
37. 10.1038/s41586-022-05592-210.1038/s41586-022-05592-2 ↗
38. 10.1016/j.cej.2024.15425210.1016/j.cej.2024.154252 ↗
39. 10.1038/s41467-025-63636-310.1038/s41467-025-63636-3 ↗
40. 10.1126/sciadv.abl374210.1126/sciadv.abl3742 ↗
41. Y.Lin S.Han H.Mao Y.Wang W. J.Dally arXiv preprintarXiv:1712.018872017.
42. M.Horowitz presented atIEEE International Solid‐State Circuits Conference Digest of Technical Papers (ISSCC) San Francisco CA USA February2014 p:10.10.1109/ISSCC.2014.6757323 ↗
43. Riquelme C. · Adv. Neural Inform. Process. Sys. · 2021 · 34 · 8583
44. 10.1073/pnas.231188512110.1073/pnas.2311885121 ↗
45. 10.1002/wps.2125610.1002/wps.21256 ↗
46. 10.1016/S0140-6736(21)00359-710.1016/S0140-6736(21)00359-7 ↗
47. 10.1038/s41572-019-0102-310.1038/s41572-019-0102-3 ↗
48. 10.1056/NEJMc191594310.1056/NEJMc1915943 ↗
49. 10.1038/s41574-022-00746-810.1038/s41574-022-00746-8 ↗
50. 10.1109/TII.2023.327491310.1109/TII.2023.3274913 ↗
51. 10.1109/JIOT.2024.351466610.1109/JIOT.2024.3514666 ↗
52. 10.1109/TCYB.2019.290515710.1109/TCYB.2019.2905157 ↗
53. 10.1002/pol.1966.11004031010.1002/pol.1966.110040310 ↗
54. 10.3389/fpsyt.2021.70758110.3389/fpsyt.2021.707581 ↗