Implementation of Highly Reliable and Energy Efficient in‐Memory Hamming Distance Computations in 1 Kb 1‐Transistor‐1‐Memristor Arrays

Lin, Huai; Wu, Zuheng; Liu, Long; Wang, Di; Zhao, Xiao Guang; Cheng, Lingli; Lin, Ya; Wang, Zhongqiang; Xu, Xiaoxin; Liu, Qi; Xing, Guozhong

doi:10.1002/admt.202100745

Cited by 14 publications

(13 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To reduce its computational complexity, one notable option is to use in-memory computing that executes searches in an analogue manner. It has recently been shown that the associative memory can be realized by analog in-memory computing based on crossbar arrays of emerging non-volatile memories [36][37][38][39] . Besides improving the computational density and energy efficiency, this paves the way for reducing the computational complexity of the associative memory to O(1).…”

Section: Discussionmentioning

confidence: 99%

A Neuro-vector-symbolic Architecture for Solving Raven's Progressive Matrices

Hersche¹,

Zeqiri²,

Benini³

et al. 2022

Preprint

View full text Add to dashboard Cite

Neither deep neural networks nor symbolic AI alone have approached the kind of intelligence expressed in humans. This is mainly because neural networks are not able to decompose distinct objects from their joint representation (the so-called binding problem), while symbolic AI suffers from exhaustive rule searches, among other problems. These two problems are still pronounced in neuro-symbolic AI which aims to combine the best of the two paradigms.Here, we show that the two problems can be addressed with our proposed neuro-vector-symbolic architecture (NVSA) by exploiting its powerful operators on fixed-width holographic vectorized representations that serve as a common language between neural networks and symbolic logical reasoning. The efficacy of NVSA is demonstrated by solving the Raven's progressive matrices. NVSA achieves a new record of 97.7% average accuracy in RAVEN, and 98.8% in I-RAVEN datasets, with two orders of magnitude faster execution than the symbolic logical reasoning on CPUs.Human fluid intelligence is the ability to think and reason abstractly, and make inferences in a novel domain. The Raven's progressive matrices (RPM) 1 test has been a widely-used assessment of fluid intelligence and abstract reasoning 2,3 . The RPM is a non-verbal test which involves perceiving pattern continuation, element abstraction, and finding relations between abstract elements based on underlying rules. Each RPM test is an analogy problem presented as a 3×3 pictorial matrix of context panels. Every panel in the matrix is filled with several geometric objects based on a certain rule, except the last panel which is left blank. The participants are asked to complete the missing panel in the matrix by picking the correct answer from a set of candidate answer panels that matches the implicit rule (see Methods and Supplementary Fig. 1). Solving this test mainly involves two aspects of intelligence: visual perception and abstract reasoning.As opposed to deep learning methods that blend perception and reasoning in a monolithic model [4][5][6] , the reasoning capability is not necessarily interwoven with the visual perception in humans. For instance, one can close one's eyes and build a scene representation through touch, followed by reasoning that remains effortless without the vision 7 . This decoupling is at the core of hybrid systems that advocate combining subsymbolic (e.g., neural networks) with symbolic artificial intelligence, aiming to reach human-level generalization [8][9][10][11] . Among the hybrid systems, neuro-symbolic architectures separately handle the perception and reasoning aspects by using a combination of neural networks and symbolic approaches. Considerable effort has been devoted to integrate these two paradigms that led to the state-of-the-art performance of neuro-symbolic architectures in various tasks, e.g., visual question answering 7,12,13 , causal video reasoning 14 , and solving RPM 15 . However, the resulting neuro-symbolic architectures are not immune to the potential problems of their indiv...

show abstract

Section: Discussionmentioning

confidence: 99%

A Neuro-vector-symbolic Architecture for Solving Raven's Progressive Matrices

Hersche¹,

Zeqiri²,

Benini³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Therefore, realizing OR/NOR IMC result output by V out and V out . Furthermore, based on the output of AND/NAND and OR/NOR, different combinations of logic gates can be used to further implement logic operations such as XOR and XNOR, etc [71].…”

Section: Non-volatile In-memory Computingmentioning

confidence: 99%

All-Electrical Control of Compact SOT-MRAM: Toward Highly Efficient and Reliable Non-Volatile In-Memory Computing

et al. 2022

Self Cite

View full text Add to dashboard Cite

Two-dimensional van der Waals (2D vdW) ferromagnets possess outstanding scalability, controllable ferromagnetism, and out-of-plane anisotropy, enabling the compact spintronics-based non-volatile in-memory computing (nv-IMC) that promises to tackle the memory wall bottleneck issue. Here, by employing the intriguing room-temperature ferromagnetic characteristics of emerging 2D Fe3GeTe2 with the dissimilar electronic structure of the two spin-conducting channels, we report on a new type of non-volatile spin-orbit torque (SOT) magnetic tunnel junction (MTJ) device based on Fe3GeTe2/MgO/Fe3GeTe2 heterostructure, which demonstrates the uni-polar and high-speed field-free magnetization switching by adjusting the ratio of field-like torque to damping-like torque coefficient in the free layer. Compared to the conventional 2T1M structure, the developed 3-transistor-2-MTJ (3T2M) cell is implemented with the complementary data storage feature and the enhanced sensing margin of 201.4% (from 271.7 mV to 547.2 mV) and 276% (from 188.2 mV to 520 mV) for reading “1” and “0”, respectively. Moreover, superior to the traditional CoFeB-based MTJ memory cell counterpart, the 3T2M crossbar array architecture can be executed for AND/NAND, OR/NOR Boolean logic operation with a fast latency of 24 ps and ultra-low power consumption of 2.47 fJ/bit. Such device to architecture design with elaborated micro-magnetic and circuit-level simulation results shows great potential for realizing high-performance 2D material-based compact SOT magnetic random-access memory, facilitating new applications of highly reliable and energy-efficient nv-IMC.

show abstract

“…Emerging nonvolatile memory (eNVM)‐based IMCs, owing to their advantageous energy efficiency and compatibility with CMOS process integration, have been extensively investigated, [ 5–11 ] including resistive RAM (RRAM, a.k.a. memristor), [ 12,13 ] phase‐change memory (PCM), [ 14 ] ferroelectric RAM (FeRAM) or tunnel junction (FTJ), [ 15,16 ] multiferroic material‐based resistive memory, [ 17 ] and spin‐transfer/spin–orbit torque (STT/SOT)‐magnetic random access memory (MRAM). [ 18–20 ] Particularly, SOT‐MRAM is considered as a promising candidate for IMC as it demonstrates competitive switching speeds (subnanosecond, ns) and almost infinite endurance cycles owing to the decoupled write and read paths.…”

Section: Introductionmentioning

confidence: 99%

Implementation of Highly Reliable and Energy‐Efficient Nonvolatile In‐Memory Computing using Multistate Domain Wall Spin–Orbit Torque Device

Lin

Wang

et al. 2022

Advanced Intelligent Systems

Self Cite

View full text Add to dashboard Cite

To achieve high computing throughputs for data-intensive tasks in real time (e.g., autonomous driving, virtual reality, and deep learning), the development of inmemory computing (IMC) architectures is emerging as a research surge. [1,2] The key concept of IMC is the implementation of arithmetic logic units in memorybased hardware to better exploit memory bandwidth and substantially improve energy efficiency during data migration. Consequently, dynamic random access memory (DRAM)-and static random access memory (SRAM)-based IMC architectures have been widely reported. [3,4] However, these designs suffer inevitable shortcomings, such as in high standby power and long latencies caused by necessary write-back operations, which hinder their potential. Emerging nonvolatile memory (eNVM)-based IMCs, owing to their advantageous energy efficiency and compatibility with

show abstract

Implementation of Highly Reliable and Energy Efficient in‐Memory Hamming Distance Computations in 1 Kb 1‐Transistor‐1‐Memristor Arrays

Cited by 14 publications

References 53 publications

A Neuro-vector-symbolic Architecture for Solving Raven's Progressive Matrices

A Neuro-vector-symbolic Architecture for Solving Raven's Progressive Matrices

All-Electrical Control of Compact SOT-MRAM: Toward Highly Efficient and Reliable Non-Volatile In-Memory Computing

Implementation of Highly Reliable and Energy‐Efficient Nonvolatile In‐Memory Computing using Multistate Domain Wall Spin–Orbit Torque Device

Contact Info

Product

Resources

About