Recent Advances and Future Prospects for Memristive Materials, Devices, and Systems

Abstract

Memristive technology has been rapidly emerging as a potential alternative to traditional CMOS technology, which is facing fundamental limitations in its development. Since oxide-based resistive switches were demonstrated as memristors in 2008, memristive devices have garnered significant attention due to their biomimetic memory properties, which promise to significantly improve power consumption in computing applications. Here, we provide a comprehensive overview of recent advances in memristive technology, including memristive devices, theory, algorithms, architectures, and systems. In addition, we discuss research directions for various applications of memristive technology including hardware accelerators for artificial intelligence, in-sensor computing, and probabilistic computing. Finally, we provide a forward-looking perspective on the future of memristive technology, outlining the challenges and opportunities for further research and innovation in this field. By providing an up-to-date overview of the state-of-the-art in memristive technology, this review aims to inform and inspire further research in this field.

2D Semiconductor‐Based Optoelectronics for Artificial Vision

“…[19,146] The artificial optoelectronic synapse could be realized in twoterminal memristors and three-terminal transistors. [93,147,148] He et al reported a W/n-MoS 2 /p-Si, demonstrating diverse synaptic neuromorphic functions, such as potentiation/habituation, long-term/short-term memory and paired-pulse facilitation. [149] In three-terminal devices, charge-trap transistors, floating-gate transistors and ferroelectric transistors are three type structure of optoelectronic synapse.…”

Reliable Memristive Synapses Based on Parylene-MoOx Nanocomposites for Neuromorphic Applications

“…The reason lies in the stochastic nature of the RS in the materials used in memristors . Thus, finding optimal materials to fabricate analog memristors suitable for use in NCSs is a critical development area in neuromorphic computing …”

“…Researchers are also exploring new memristive or quasi-memristive structures influenced by external stimuli other than electrical ones, such as magnetic fields, light, or temperature . However, the “classical” sandwich memristors, driven by an electric field, remain pivotal in the discipline and are widely viewed as one of the most promising for microelectronics …”

“…15,16 Memristors can also be utilized in other computational tasks, including in-sensor and probabilistic computing. 17 The critical parameters for using memristors in NCSs include the number of stable (without degradation) cycles of RS from a high-resistance state (the HRS, R off ) to a low-resistance one (the LRS, R on ) and vice versa, the R off /R on ratio, resistive plasticity (or the number of stable resistive states), retention of the states, power consumption, and electroforming-free switching. 18,19 Numerous studies have shown that it is currently not possible to achieve optimal values for all of these parameters at once.…”

“…21 Thus, finding optimal materials to fabricate analog memristors suitable for use in NCSs is a critical development area in neuromorphic computing. 17 Memristors typically have a sandwich structure of a metal/RS layer/metal. Efforts are ongoing to find a suitable material for the RS layer to achieve optimal memristive characteristics.…”