High-Throughput, Area-Efficient, and Variation-Tolerant 3-D In-Memory Compute System for Deep Convolutional Neural Networks

Veluri, Hasita; Li, Yida; Niu, Jessie Xuhua; Zamburg, Evgeny; Thean, Aaron

doi:10.1109/jiot.2021.3058015

Cited by 16 publications

(10 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The conductance nonlinearity of many nonvolatile memories inevitably leads to the need to implement a weight update (the analog conductance modulation) with nonidentical pulse programming (NPP) because the pulse widths or amplitudes are adjusted to compensate for the nonlinear conductance change as the analog memory states change . In addition, to minimize the data encoding (quantization) error, the NPP mode requires high-resolution digital-to-analog converters (DACs) to produce a range of pulse amplitudes, leading to increased energy consumption and area penalties in circuits . In contrast, an ideal analog memory provides a linear conductance change with identical pulse programming (IPP).…”

Section: Results and Discussionmentioning

confidence: 99%

“…Applying in-memory computing to address the challenges of the central processing unit (CPU) latency and energy constraints of memory access is gaining attention for next-generation computing architectures. , Ferroelectric field-effect transistors (FeFETs) are promising three-terminal devices for in-memory computing due to their analog memory states, faster operation speed, lower energy consumption, and no read–disturb issue in comparison to two terminal memristors, such as resistive random access memory . The structure of a FeFET contains a semiconductor material as the channel layer and a ferroelectric material as the gate dielectric layer.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Stress-Memorized HZO for High-Performance Ferroelectric Field-Effect Memtransistor

Tsai

Fang

Wang

et al. 2022

ACS Appl. Electron. Mater.

Self Cite

View full text Add to dashboard Cite

A ferroelectric field-effect transistor (FeFET), capable of logic and memory functionalities in a single device, is a promising three-terminal memtransistor that enables high-performance in-memory computing for non Von Neumann architectures. Among all HfO2-based ferroelectric materials, HfZrO2 (HZO) has attracted the most attention due to the low process temperature of ≤500 °C; however, it has relatively weak polarization. Many prior works claimed that the way to improve HZO-based FeFET characteristics is to enhance HZO ferroelectric properties, while they did not account for the fundamental compromise on dielectric breakdown strength (BDS), transistor ON/OFF current (I ON/I OFF) ratio, and memory window (MW) due to the enhanced polarization. In this work, we propose an approach for controlling the ferroelectric orthorhombic phase (O phase) and the corresponding polarization in optimal value by engineering both the surface morphology and stress of HZO layer by a thermal expansion mismatch with a TiN/W stacked capping layer, to improve the BDS, I ON/I OFF ratio, and MW. Through electrode surface optimization and stress memorization we achieved an 18% HZO ferroelectricity increase with a high BDS value of ≥4.8 MV/cm. Our optimized FeFET shows good electrical characteristics and supports operation in an identical pulse programming (IPP) mode, showing good potentiation and depression nonlinearity (−0.84 and −2.04) with an asymmetry factor of 1.2. A simulation based on the proposed FeFET array demonstrates the high potential of application in an artificial neural network (ANN).

show abstract

Section: Results and Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Stress-Memorized HZO for High-Performance Ferroelectric Field-Effect Memtransistor

Tsai

Fang

Wang

et al. 2022

ACS Appl. Electron. Mater.

Self Cite

View full text Add to dashboard Cite

show abstract

“…At the moment, the main efforts of researchers and engineers are focused on the co-optimization of memristive materials and devices in accordance with the requirements for specific emerging systems and technologies. Three-dimensional integration of memristive devices is a promising way to the development of future ultralarge neuromorphic integrated circuits that can approach the capabilities of the human brain (Veluri, Li, Niu, Zamburg, & Thean, 2021). An important role on this way is played by the implementation of a systematic approach to the development of the entire chain of computer-aided design (CAD) tools from devices to algorithms and hybrid software-hardware simulation systems discussed in Section 3.3.…”

Section: Hardware Implementationmentioning

confidence: 99%

Neuropunk Revolution. Hacking Cognitive Systems towards Cyborgs 3.0

Talanov¹,

Vallverdú²,

Adamatzky³

et al. 2022

Preprint

View full text Add to dashboard Cite

This work is dedicated to the review and perspective of the new direction that we call "Neuropunk revolution" resembling the cultural phenomenon of cyberpunk. This new phenomenon has its foundations in advances in neuromorphic technologies including memristive and bio-plausible simulations, BCI, and neurointerfaces as well as unconventional approaches to AI and computing in general. We present the review of the current state-of-the-art and our vision of near future development of scientific approaches and future technologies. We call the "Neuropunk revolution" the set of trends that in our view provide the necessary background for the new generation of approaches technologies to integrate the cybernetic objects with biological tissues in close loop system as well as robotic systems inspired by the biological processes again integrated with biological objects. We see bio-plausible simulations implemented by digital computers or spiking networks memristive hardware as promising bridge or middleware between digital and [neuro]biological domains.

show abstract

“…To break the so-called “von Neumann bottleneck”, , adjacent or syncretic architectures of memory and processor are being explored, such as memory hierarchy, − near-memory computation, , and in-memory computing. , In-memory computing, which is analogous to the biological brain, , utilizes in situ memory and computation to dramatically improve energy and processing efficiency for data-intensive and complex logic applications. − …”

Section: Introductionmentioning

confidence: 99%

Van der Waals Ferroelectric Semiconductor Field Effect Transistor for In-Memory Computing

Liao

Wen

et al. 2023

ACS Nano

View full text Add to dashboard Cite

In-memory computing is a highly efficient approach for breaking the bottleneck of von Neumann architectures, i.e., reducing redundant latency and energy consumption during the data transfer between the physically separated memory and processing units. Herein we have designed a in-memory computing device, a van der Waals ferroelectric semiconductor (InSe) based metal-oxide-ferroelectric semiconductor field-effect transistor (MOfeS-FET). This MOfeS-FET integrates memory and logic functions in the same material, in which the out-of-plane (OOP) ferroelectric polarization in InSe is used for data storage and the semiconducting property is used for the logic computation. The MOfeS-FET shows a long retention time with high on/off ratios (>10 6 ), high program/erase (P/E) ratios (10 3 ), and stable cyclic endurance. Moreover, inverter, programmable NAND, and NOR Boolean logic operations with nonvolatile storage of the results have all been demonstrated using our approach. These findings highlight the potential of van der Waals ferroelectric semiconductor-based MOfeS-FETs in the in-memory computing and their potential of achieving size scaling beyond Moore's law.

show abstract

High-Throughput, Area-Efficient, and Variation-Tolerant 3-D In-Memory Compute System for Deep Convolutional Neural Networks

Cited by 16 publications

References 46 publications

Stress-Memorized HZO for High-Performance Ferroelectric Field-Effect Memtransistor

Stress-Memorized HZO for High-Performance Ferroelectric Field-Effect Memtransistor

Neuropunk Revolution. Hacking Cognitive Systems towards Cyborgs 3.0

Van der Waals Ferroelectric Semiconductor Field Effect Transistor for In-Memory Computing

Contact Info

Product

Resources

About