Abstract: We propose a three-neuron heterogeneous cyclic Hopfield neural network (het-CHNN) utilizing three different activation functions: the hyperbolic tangent, sine, and cosine functions. The network’s globally uniformly ultimate boundedness is proved theoretically, and its chaotic dynamics are explored through numerical simulations and analog experiments. The numerical results demonstrate that the het-CHNN displays chaotic dynamics and multi-scroll chaotic attractors. Subsequently, the het-CHNN is implemented in an analog circuit, and hardware experiments are performed to verify the previous numerical results. Notably, the het-CHNN successfully resolves the issue of the absence of chaos in a three-neuron CHNN and currently appears to be the simplest three-neuron Hopfield neural network (HNN) that can generate chaos.

一种无自连接的异构循环型霍普菲尔德神经网络

[1]BaoBC, ChenCJ, BaoH, et al., 2019. Dynamical effects of neuron activation gradient on Hopfield neural network: numerical analyses and hardware experiments. International Journal of Bifurcation and Chaos, 29(4):1930010.

[2]BaoBC, TangHG, SuYH, et al., 2024. Two-dimensional discrete bi-neuron Hopfield neural network with polyhedral hyperchaos. IEEE Transactions on Circuits and Systems I: Regular Papers, 71(12):5907-5918.

[3]BaoH, ChenZG, CaiJM, et al., 2022. Memristive cyclic three-neuron-based neural network with chaos and global coexisting attractors. Science China Technological Sciences, 65(11):2582-2592.

[4]BaoH, ChenZG, MaJ, et al., 2024. Planar homogeneous coexisting hyperchaos in bi-memristor cyclic Hopfield neural network. IEEE Transactions on Industrial Electronics, 71(12):16398-16408.

[5]ChenB, XuQ, ChenM, et al., 2021. Initial-condition-switched boosting extreme multistability and mechanism analysis in a memcapacitive oscillator. Frontiers of Information Technology & Electronic Engineering, 22(11):1517-1531.

[6]ChenCJ, MinFH, CaiJM, et al., 2024. Memristor synapse-driven simplified Hopfield neural network: hidden dynamics, attractor control, and circuit implementation. IEEE Transactions on Circuits and Systems I: Regular Papers, 71(5):2308-2319.

[7]ChenM, RenX, WuHG, et al., 2019. Periodically varied initial offset boosting behaviors in a memristive system with cosine memductance. Frontiers of Information Technology & Electronic Engineering, 20(12):1706-1716.

[8]ChenM, RenX, WuHG, et al., 2020. Interpreting initial offset boosting via reconstitution in integral domain. Chaos, Solitons & Fractals, 131:109544.

[9]DancaMF, KuznetsovN, 2017. Hidden chaotic sets in a Hopfield neural system. Chaos, Solitons & Fractals, 103:144-150.

[10]DingDW, ChenSQ, ZhangHW, et al., 2024. Firing pattern transition of fractional-order memristor-coupled Hindmarsh-Rose neurons model and its medical image encryption for region of interest. Nonlinear Dynamics, 112(12):10529-10554.

[11]GratwickeJ, JahanshahiM, FoltynieT, 2015. Parkinson’s disease dementia: a neural networks perspective. Brain, 138(6):1454-1476.

[12]HopfieldJJ, 1982. Neural networks and physical systems with emergent collective computational abilities. Proceedings of the National Academy of Sciences of the United States of America, 79(8):2554-2558.

[13]KhalilHK, 2002. Nonlinear Systems. 3rd Edition. Prentice Hall, Upper Saddle River, USA.

[14]KobayashiM, 2020. Diagonal rotor Hopfield neural networks. Neurocomputing, 415:40-47.

[15]KornH, FaureP, 2003. Is there chaos in the brain? II. Experimental evidence and related models. Comptes Rendus Biologies, 326(9):787-840.

[16]LaiQ, WanZQ, KuatePDK, 2023. Generating grid multi-scroll attractors in memristive neural networks. IEEE Transactions on Circuits and Systems I: Regular Papers, 70(3):1324-1336.

[17]LaiQ, YangL, HuGW, et al., 2024. Constructing multiscroll memristive neural network with local activity memristor and application in image encryption. IEEE Transactions on Cybernetics, 54(7):4039-4048.

[18]LeiZ, GuoQ, WangCN, et al., 2025. Continuous energy exchange between magnetic fields supporting memristive neuron firing. Journal of Zhejiang University-SCIENCE A, 26(8):755-770.

[19]LiFY, ChenZG, ZhangYZ, et al., 2024a. Cascade tri-neuron Hopfield neural network: dynamical analysis and analog circuit implementation. AEÜ-International Journal of Electronics and Communications, 174:155037.

[20]LiFY, BaiLF, ChenZG, et al., 2024b. Scroll-growth and scroll-control attractors in memristive bi-neuron Hopfield neural network. IEEE Transactions on Circuits and Systems II: Express Briefs, 71(4):2354-2358.

[21]LiFY, QinWS, XiMQ, et al., 2025. Plane coexistence behaviors for Hopfield neural network with two-memristor-interconnected neurons. Neural Networks, 183:107049.

[22]LinHR, WangCH, CuiL, et al., 2022. Brain-like initial-boosted hyperchaos and application in biomedical image encryption. IEEE Transactions on Industrial Informatics, 18(12):8839-8850.

[23]LinHR, WangCH, XuC, et al., 2023. A memristive synapse control method to generate diversified multistructure chaotic attractors. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 42(3):942-955.

[24]LinW, ChenGR, 2009. Large memory capacity in chaotic artificial neural networks: a view of the anti-integrable limit. IEEE Transactions on Neural Networks, 20(8):1340-1351.

[25]LiuL, HuangY, ChenZG, et al., 2025. A dual-neuron memristive Hopfield neural network and its application in image encryption. Nonlinear Dynamics, 113(14):18705-18726.

[26]MaJ, 2023. Biophysical neurons, energy, and synapse controllability: a review. Journal of Zhejiang University-SCIENCE A, 24(2):109-129.

[27]McFarlanAR, ChouCYC, WatanabeA, et al., 2023. The plasticitome of cortical interneurons. Nature Reviews Neuroscience, 24(2):80-97.

[28]NjitackeZT, IsaacSD, NestorT, et al., 2021. Window of multistability and its control in a simple 3D Hopfield neural network: application to biomedical image encryption. Neural Computing and Applications, 33(12):6733-6752.

[29]PrescottSL, LiberlesSD, 2022. Internal senses of the vagus nerve. Neuron, 110(4):579-599.

[30]RechPC, 2015. Period-adding and spiral organization of the periodicity in a Hopfield neural network. International Journal of Machine Learning and Cybernetics, 6(1):1-6.

[31]SilvaCP, 1993. Shil’nikov’s theorem-a tutorial. IEEE Transactions on Circuits and Systems I: Fundamental Theory and Applications, 40(10):675-682.

[32]TangD, WangCH, LinHR, et al., 2024. Dynamics analysis and hardware implementation of multi-scroll hyperchaotic hidden attractors based on locally active memristive Hopfield neural network. Nonlinear Dynamics, 112(2):1511-1527.

[33]WanQZ, LiF, ChenSM, et al., 2023. Symmetric multi-scroll attractors in magnetized Hopfield neural network under pulse controlled memristor and pulse current stimulation. Chaos, Solitons & Fractals, 169:113259.

[34]WangCH, LiangJH, DengQL, 2024. Dynamics of heterogeneous Hopfield neural network with adaptive activation function based on memristor. Neural Networks, 178:106408.

[35]WangN, LiCQ, BaoH, et al., 2019. Generating multi-scroll Chua’s attractors via simplified piecewise-linear Chua’s diode. IEEE Transactions on Circuits and Systems I: Regular Papers, 66(12):4767-4779.

[36]XieY, YaoZ, MaJ, 2022. Phase synchronization and energy balance between neurons. Frontiers of Information Technology & Electronic Engineering, 23(9):1407-1420.

[37]XuSC, WangXY, YeXL, 2022. A new fractional-order chaos system of Hopfield neural network and its application in image encryption. Chaos, Solitons & Fractals, 157:111889.

[38]YangXS, 2008. 3-D cellular neural networks with cyclic connections cannot exhibit chaos. International Journal of Bifurcation and Chaos, 18(4):1227-1230.

[39]YuF, ShenH, YuQL, et al., 2023. Privacy protection of medical data based on multi-scroll memristive Hopfield neural network. IEEE Transactions on Network Science and Engineering, 10(2):845-858.

[40]YuF, LinY, YaoW, et al., 2025a. Multiscroll Hopfield neural network with extreme multistability and its application in video encryption for IIoT. Neural Networks, 182:106904.

[41]YuF, SuD, HeSQ, et al., 2025b. Resonant tunneling diode cellular neural network with memristor coupling and its application in police forensic digital image protection. Chinese Physics B, 34(5):050502.

[42]ZhangM, EichhornSW, ZinggB, et al., 2021. Spatially resolved cell atlas of the mouse primary motor cortex by MERFISH. Nature, 598(7879):137-143.

[43]ZhangS, ChenCJ, ZhangYZ, et al., 2025. Multidirectional multidouble-scroll Hopfield neural network with application to image encryption. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 55(1):735-746.

[44]ZhouCL, BaoH, ZhangYZ, et al., 2026. Memory capacity expansion in a sine activated Hopfield neural network. Chaos, Solitons & Fractals, 205:117843.