Search for: All records

Edge devices face challenges when implementing deep neural networks due to constraints on their computational resources and power consumption. Fuzzy logic systems can potentially provide more efficient edge implementations due to their compactness and capacity to manage uncertain data. However, their hardware realization remains difficult, primarily because implementing reconfigurable membership function generators using conventional technologies requires high circuit complexity and power consumption. Here we report a multigate van der Waals interfacial junction transistor based on a molybdenum disulfide/graphene heterostructure that can generate tunable Gaussian-like and π-shaped membership functions. By integrating these generators with peripheral circuits, we create a reconfigurable fuzzy controller hardware capable of nonlinear system control. This fuzzy logic system can also be integrated with a few-layer convolution neural network to form a fuzzy neural network with enhanced performance in image segmentation. 

Abstract Phase change materials, which show different electrical characteristics across the phase transitions, have attracted considerable research attention for their potential electronic device applications. Materials with metal‐to‐insulator or charge density wave (CDW) transitions such as VO₂and 1T‐TaS₂have demonstrated voltage oscillations due to their robust bi‐state resistive switching behavior with some basic neuronal characteristics. BaTiS₃is a small bandgap ternary chalcogenide that has recently reported the emergence of CDW order below 245 K. Here, the discovery of DC voltage / current‐induced reversible threshold switching in BaTiS₃devices between a CDW phase and a room temperature semiconducting phase is reported. The resistive switching behavior is consistent with a Joule heating scheme and sustained voltage oscillations with a frequency of up to 1 kHz are demonstrated by leveraging the CDW phase transition and the associated negative differential resistance. Strategies of reducing channel sizes and improving thermal management may further improve the device's performance. The findings establish BaTiS₃as a promising CDW material for future electronic device applications, especially for energy‐efficient neuromorphic computing. 

Abstract Synaptic devices with tunable weight hold great promise in enabling non-von Neumann architecture for energy efficient computing. However, conventional metal-insulator-metal based two-terminal memristors share the same physical channel for both programming and reading, therefore the programming power consumption is dependent on the synaptic resistance states and can be particularly high when the memristor is in the low resistance states. Three terminal synaptic transistors, on the other hand, allow synchronous programming and reading and have been shown to possess excellent reliability. Here we present a binary oxide based three-terminal MoS₂synaptic device, in which the channel conductance can be modulated by interfacial charges generated at the oxide interface driven by Maxwell-Wagner instability. The binary oxide stack serves both as an interfacial charge host and gate dielectrics. Both excitatory and inhibitory behaviors are experimentally realized, and the presynaptic potential polarity can be effectively controlled by engineering the oxide stacking sequence, which is a unique feature compared with existing charge-trap based synaptic devices and provides a new tuning knob for controlling synaptic device characteristics. By adopting a three-terminal transistor structure, the programming channel and reading channel are physically separated and the programming power consumption can be kept constantly low (∼50 pW) across a wide dynamic range of 10⁵. This work demonstrates a complementary metal oxide semiconductor compatible approach to build power efficient synaptic devices for artificial intelligence applications. 

Memristive devices can offer dynamic behaviour, analogue programmability, and scaling and integration capabilities. As a result, they are of potential use in the development of information processing and storage devices for both conventional and unconventional computing paradigms. Their memristive switching processes originate mainly from the modulation of the number and position of structural defects or compositional impurities—what are commonly referred to as imperfections. While the underlying mechanisms and potential applications of memristors based on traditional bulk materials have been extensively studied, memristors based on van der Waals materials have only been considered more recently. Here we examine imperfection-enabled memristive switching in van der Waals materials. We explore how imperfections— together with the inherent physicochemical properties of the van der Waals materials—create different switching mechanisms, and thus provide a range of opportunities to engineer switching behaviour in memristive devices. We also discuss the challenges involved in terms of material selection, mechanism investigation and switching uniformity control, and consider the potential of van der Waals memristors in system-level implementations of efficient computing technologies.