Hardware-backpropagation learning of neuron MOS neural networks

Large-Area Soft e-Skin: The Challenges Beyond Sensor Designs

“…The υ-NWFETsbased circuits could also exhibit similar plastic behavior, as discussed in [48]. Other hardware neuromorphic architecture implementations reported in the literature are based on devices such as memristor [90], spin logic [91], [92], memristor [93], neuron MOSFET [88], [94], analog circuit-based neurons [95], field-programmable gate array (FPGA) [75], and software-programmed NNs [96]. Although none of these have been used for tactile skin, they could possibly offer solutions.…”

“…7(a)] in a simplified manner, can be an excellent building block for such hardware NNs (HNNs) [48]. The structure of υ-NWFET is a variant of a neuron MOSFET with NWs providing the functional channel region [86]- [88]. The floating gate of υ-NWFET, which modulates the channel current, is capacitively coupled to several gates.…”

Neuron firing operations by a new basic logic element

“…N EURAL logic devices and circuits have been extensively studied [1], [2], and recently, a MOS transistor and its logic circuits were proposed [3], [4]. While quantum effect devices may contribute to new functional applications [5], they are still somewhat immature in terms of practical application.…”

Hardware-learning neural network LSI using a highly-functional transistor simulating neuron actions

“…In addition, a new learning algorithm called Hardware Backpropagation (HBP) has been also developed in order to facilitate the hardware implementation of self-learning functions [9]. Based on this algorithm, the actions for self-learning are also conducted in voltage-mode by a digital-analog merged YMOS circuit technology.…”

“…The very powerful functional capability of YMOS has been successfully applied to implementing binary logic circuits, resulting in a dramatic reduction in the transistor count as well as interconnections [4-71. Since a vMOS transistor imitates the action of biological neurons at the very primitive cell-level function, the synthesis of neural network is the most natural extension of YMOS circuits applications [8,9].…”