Mushroom-Type phase change memory with projection liner: An array-level demonstration of conductance drift and noise mitigation

Bruce, Robert L.; Sarwat, Syed Ghazi; Boybat, Irem; Cheng, C.-W.; Kim, W.; Nandakumar, S. R.; Mackin, Charles; Philip, Timothy M.; Liu, Z.; Brew, Kevin; Gong, N.; Ok, I.; Adusumilli, Praneet; Spoon, Katherine; Ambrogio, Stefano; Kersting, Benedikt; Bohnstingl, Thomas; Gallo, Manuel Le; Simon, A.; Li, N.; Saraf, Iqbal; Han, Jin‐Ping; Gignac, Lynne; Papalia, John M.; Yamashita, Toshiro; Saulnier, Nicole; Burr, Geoffrey W.; Tsai, Hsinyu; Sebastian, Abu; Narayanan, V.; BrightSky, M.

doi:10.1109/irps46558.2021.9405191

Cited by 15 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of physiological processing signals, [96] MLP RRAM MNIST --BP Q. Zhang, et al [97] MLP RRAM MNIST --SBP I. Giannopoulos, et al [100] MLP PCM 4/1/0 classification 30 B --S. Oh, et al [98] MLP PCM MNIST ---X. Sun, et al [99] MLP PCM CIFAR-10 -40 nm WT D. Garbin, et al [106] CNN RRAM MNIST -65 nm -T. Gokmen, et al [107] CNN RRAM MNIST --BP Z. Dong, et al [102] CNN RRAM MNIST 128 B -STDP M. Davies, et al [118] CNN RRAM CIFAR-10 256 MB 45 nm -V. Joshi, et al [108] CNN PCM CIFAR-10&ImageNet 1 MB 90 nm -S. Oh, et al [98] CNN PCM MNIST ---Y. Lin, et al [109] BayNN RRAM MNIST 160 kB --A. Malhotra, et al [110] BayNN RRAM MNIST ---A. G. Kusne, et al [111] BayNN PCM ICSD&AFLOW.org ---N. Gong, et al [112] BayNN PCM GPR 1 kB 90 nm -C. Li, et al [116] LSTM RRAM USF-NIST 2 kB 2 um BPTT H. Nikam, et al [114] LSTM RRAM Passenger-count Prediction 4 kB -BP H. Tsai, et al [117] LSTM PCM Alice in Wonderland 2.5 MB 90 nm -R. L. Bruce, et al [113] LSTM PCM Penn Tree Bank 1 MB 90 nm -spiking neurons have more advantages than artificial neurons such as bionics, [119] physiological signal processing, [120] and event-driven. [121]…”

Section: Spiking Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Integrated Memristor Network for Physiological Signal Processing

Cai

Yue

et al. 2023

Adv Elect Materials

View full text Add to dashboard Cite

Humans are complex organisms made by millions of physiological systems. Therefore, physiological activities can represent physical or mental states of the human body. Physiological signal processing is essential in monitoring human physiological features. For example, non‐invasive electroencephalography (EEG) signals can be used to reconstruct brain consciousness and detect eye movements for identity verification. However, physiological signal processing requires high resolution, high sensitivity, fast responses, and low power consumption, hindering practical hardware design for physiological signal processing. The bionic capability of memristor devices is very promising in the context of building physiological signal processing hardware and they have demonstrated a handful of advantages over the traditional Von Neumann architecture system in accelerating neural networks. Memristor networks can be integrated as a hardware system for physiological signal processing that can deliver higher energy efficiency and lower latency compared to traditional implementations. This review paper first introduces memristor characteristics, followed by a comprehensive literature study of memristor‐based networks. Physiology signal processing applications enabled by these integrated memristor networks are also presented in this review. In summary, this paper aims to provide a new perspective on physiological signal processing using integrated memristor networks.

show abstract

Section: Spiking Neural Networkmentioning

confidence: 99%

Section: Integrated Memristor Networkmentioning

confidence: 99%

Integrated Memristor Network for Physiological Signal Processing

Cai

Yue

et al. 2023

Adv Elect Materials

View full text Add to dashboard Cite

show abstract

“…Employing new materials, like carbon nanotubes or graphene as electrode and advanced device shield, can obviously improve the heating efficiency during write program and restrict inter-device variability, respectively. In terms of device design and structure, the specific thin conducting surfactant layers [73,74] and the phase-change heterostructure [75] have been proved to be further effective in restricting resistance drift. As for circuit and array design, differential mTnR cell structure and reference cell-based resistance tracking provide a set of feasible solution for symmetrical conductance regulation and device uniformity.…”

Section: Advances In Science and Technology To Meet Challengesmentioning

confidence: 99%

2022 roadmap on neuromorphic devices and applications research in China

Wan¹,

Wan²,

Wu³

et al. 2022

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

The data throughput in the von Neumann architecture based computing system is limited by its separated processing and memory structure and the mismatching speed between the two units. As a result, it’s quite difficult to improve the energy efficiency in conventional computing system, especially for dealing with unstructured data. Meanwhile, artificial intelligence and robotics nowadays still behave poorly in autonomy, creativity and sociality, which has been considered as the unimaginable computational requirement for sensorimotor skills. These two plights have urged the imitation and replication of the biological systems in terms of computing, sensing, and even motoring. Hence, the so-called neuromorphic system has drawn world-wide attention in recent decade, which is aimed at addressing the aforementioned needs from the mimicking of neural system. The recent development on emerging memory devices, nanotechnologies, and materials science have provided an unprecedented opportunity for this aim. This roadmap profiles the potential trend in building neuromorphic systems from the view of Chinese scientists. The content of this roadmap will cover some core topics from multidisciplinary researchers including electronics, computer science, materials, physics, and so on. The perspectives and challenges are also discussed in partly, which may serve as the guidance for the pursuing of high energy-efficient and/or bio-realistic systems that can compute, sense, and act as our human beings. This will also give birth to some excited paradigm breakers and advanced technologies in a wide spectrum of areas.

show abstract

“…Obviously, each technique comes with its own set of drawbacks, i.e. requiring a different fabrication process technology [25], a considerable area overhead associated to the AIMC unit [13], [22], reliance on accurate device models [24] or the periodic recalibration of the system [18]. By applying multiple techniques simultaneously the requirements on each of them can be relaxed, with potential reduction of the incurred cost.…”

mentioning

confidence: 99%

Combined HW/SW Drift and Variability Mitigation for PCM-Based Analog In-Memory Computing for Neural Network Applications

Antolini

Paolino

Zavalloni

et al. 2023

IEEE J. Emerg. Sel. Topics Circuits Syst.

View full text Add to dashboard Cite

Matrix-Vector Multiplications (MVMs) represent a heavy workload for both training and inference in Deep Neural Networks (DNNs) applications. Analog In-memory Computing (AIMC) systems based on Phase Change Memory (PCM) has been shown to be a valid competitor to enhance the energy efficiency of DNN accelerators. Although DNNs are quite resilient to computation inaccuracies, PCM non-idealities could strongly affect MVM operations precision, and thus the accuracy of DNNs. In this paper, a combined hardware and software solution to mitigate the impact of PCM non-idealities is presented. The drift of PCM cells conductance is compensated at the circuit level through the introduction of a conductance ratio at the core of the MVM computation. A model of the behaviour of PCM cells is employed to develop a device-aware training for DNNs and the accuracy is estimated in a CIFAR-10 classification task. This work is supported by a PCM-based AIMC prototype, designed in a 90-nm STMicroelectronics technology, and conceived to perform Multiply-and-Accumulate (MAC) computations, which are the kernel of MVMs. Results show that the MAC computation accuracy is around 95% even under the effect of cells drift. The use of a device-aware DNN training makes the networks less sensitive to weight variability, with a 15% increase in classification accuracy over a conventionally-trained Lenet-5 DNN, and a 36% gain when drift compensation is applied.

show abstract

Mushroom-Type phase change memory with projection liner: An array-level demonstration of conductance drift and noise mitigation

Cited by 15 publications

References 10 publications

Integrated Memristor Network for Physiological Signal Processing

Integrated Memristor Network for Physiological Signal Processing

2022 roadmap on neuromorphic devices and applications research in China

Combined HW/SW Drift and Variability Mitigation for PCM-Based Analog In-Memory Computing for Neural Network Applications

Contact Info

Product

Resources

About