论文标题
神经形态的约瑟夫森连接的动力学特性
Dynamical properties of neuromorphic Josephson junctions
论文作者
论文摘要
神经形态计算利用了许多物理系统和神经元生物物理学之间的动态类比。尤其是超导系统是神经形态设备的出色候选者，因为它们的能力以极高的速度运行，并且与硅相比具有低耗能的能力。在这项研究中，我们重新审视了基于约瑟夫森连接的“神经元”的先前工作，以确定系统的神经元样特性的确切动力学机制，并揭示了与神经功能和超导神经型神经型神经功能设计有关的新复杂行为。我们的工作在于超导物理学和理论神经科学的交集，这两者都在共同的框架下，即非线性动力学理论。
Neuromorphic computing exploits the dynamical analogy between many physical systems and neuron biophysics. Superconductor systems, in particular, are excellent candidates for neuromorphic devices due to their capacity to operate in great speeds and with low energy dissipation compared to their silicon counterparts. In this study we revisit a prior work on Josephson Junction-based "neurons" in order to identify the exact dynamical mechanisms underlying the system's neuron-like properties and reveal new complex behaviors which are relevant for neurocomputation and the design of superconducting neuromorphic devices. Our work lies at the intersection of superconducting physics and theoretical neuroscience, both viewed under a common framework, that of nonlinear dynamics theory.
