A Monolithic 3D Integration of RRAM Array with Oxide Semiconductor FET for In-Memory Computing in Quantized Neural Network AI Applications
Jixuan Wu, Fei Mo, Takuya Saraya, Toshiro Hiramoto, Masaharu Kobayashi
Abstract
We have monolithically integrated RRAM array with oxide semiconductor channel access transistor in 3D stack, achieved uniform memory characteristics of 1T1R cells at each layer, and demonstrated basic functionality of XNOR operation as in-memory computing for binary neural network AI applications, for the first time. The impact of RRAM bit error rate on neural network is also investigated. 3D neural network built by this architecture has high potential to enable area-efficient, low-power and low-latency computing.
Topics & Concepts
Resistive random-access memoryArtificial neural networkXNOR gateComputer scienceNon-volatile memoryMemristorTransistorBinary numberIn-Memory ProcessingStack (abstract data type)Neuromorphic engineeringElectronic engineeringMaterials scienceComputer hardwareChannel (broadcasting)Electrical engineeringEngineeringComputer networkArtificial intelligenceNAND gateVoltageSearch engineQuery by ExampleInformation retrievalArithmeticMathematicsWeb search queryProgramming languageAdvanced Memory and Neural ComputingFerroelectric and Negative Capacitance DevicesNeural Networks and Applications