Abstract:The increasing complexity of deep learning systems has pushed conventional computing technologies to their limits. While the memristor is one of the prevailing technologies for deep learning acceleration, it is only suited for classical learning layers where only two operands, namely weights and inputs, are processed simultaneously. Meanwhile, to improve the computational efficiency of deep learning for emerging applications, a variety of non-traditional layers requiring concurrent processing of many operands are becoming popular. For example, hypernetworks improve their predictive robustness by simultaneously processing weights and inputs against the application context. Two-electrode memristor grids cannot directly map emerging layers’ higher-order multiplicative neural interactions. Addressing this unmet need, we present crossbar processing using dual-gated memtransistors based on two-dimensional semiconductor MoS 2 . Unlike the memristor, the resistance states of memtransistors can be persistently programmed and can be actively controlled by multiple gate electrodes. Thus, the discussed memtransistor crossbar enables several advanced inference architectures beyond a conventional passive crossbar. For example, we show that sneak paths can be effectively suppressed in memtransistor crossbars, whereas they limit size scalability in a passive memristor crossbar. Similarly, exploiting gate terminals to suppress crossbar weights dynamically reduces biasing power by ∼20% in memtransistor crossbars for a fully connected layer of AlexNet. On emerging layers such as hypernetworks, collocating multiple operations within the same crossbar cells reduces operating power by ∼ 15 × on the considered network cases.

A new crossbar architecture based on two serial memristors with threshold

Aging Aware Retraining for Memristor-based Neuromorphic Computing

Hybrid crossbar architecture for a memristor based memory

Crossbar architecture based on 2R complementary resistive switching memory cell

A Multi-Value 3D Crossbar Array Nonvolatile Memory Based on Pure Memristors

Design of Heterogeneous Memristor Based 1T2M Multi-value Memory Crossbar Array

Future Large-Scale Memristive Device Crossbar Arrays: Limits Imposed by Sneak-Path Currents on Read Operations.

Memristive Crossbar Arrays for Storage and Computing Applications

Design and Analysis of Crossbar Architecture Based on Complementary Resistive Switching Non-Volatile Memory Cells.

Modeling and Analysis of a Large-Scale Memristor Crossbar Array for Neuromorphic Chip

A Serial Access Scheme Design on Memristor-CMOS Hybrid Memory

Memristor Crossbars with 4.5 Terabits-per-Inch-Square Density and Two Nanometer Dimension

4K-Memristor Analog-Grade Passive Crossbar Circuit

Threshold-type memristor-based crossbar array design and its application in handwritten digit recognition

A Novel Design for Memristor-Based Logic Switch and Crossbar Circuits

One-Step Sneak-Path Free Read Scheme for Resistive Crossbar Memory

Multi-Level Programming Of Memristor In Nanocrossbar

Higher order neural processing with input-adaptive dynamic weights on MoS2 memtransistor crossbars

Post-Moore Memory Technology: Sneak Path Current (SPC) Phenomena on RRAM Crossbar Array and Solutions

A Read and Write Method for Forgetting Memristor Crossbar Array with Long-term and Short-term Memory

Read Operation Performance of Large Selectorless Cross-Point Array with Self-Rectifying Memristive Device