Neuromorphic Computing and Engineering最新文献

{"title":"Effect of conductance linearity of Ag-chalcogenide CBRAM synaptic devices on the pattern recognition accuracy of an analog neural training accelerator","authors":"P. Apsangi, H. Barnaby, M. Kozicki, Y. Gonzalez-Velo, J. Taggart","doi":"10.1088/2634-4386/ac6534","DOIUrl":"https://doi.org/10.1088/2634-4386/ac6534","url":null,"abstract":"Pattern recognition using deep neural networks (DNN) has been implemented using resistive RAM (RRAM) devices. To achieve high classification accuracy in pattern recognition with DNN systems, a linear, symmetric weight update as well as multi-level conductance (MLC) behavior of the analog synapse is required. Ag-chalcogenide based conductive bridge RAM (CBRAM) devices have demonstrated multiple resistive states making them potential candidates for use as analog synapses in neuromorphic hardware. In this work, we analyze the conductance linearity response of these devices to different pulsing schemes. We have demonstrated an improved linear response of the devices from a non-linearity factor of 6.65 to 1 for potentiation and −2.25 to −0.95 for depression with non-identical pulse application. The effect of improved linearity was quantified by simulating the devices in an artificial neural network. The classification accuracy of two-layer neural network was seen to be improved from 85% to 92% for small digit MNIST dataset.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128911729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"On-chip learning of a domain-wall-synapse-crossbar-array-based convolutional neural network","authors":"Varun Bhavin Desai, Divya Kaushik, Janak Sharda, D. Bhowmik","doi":"10.1088/2634-4386/ac62db","DOIUrl":"https://doi.org/10.1088/2634-4386/ac62db","url":null,"abstract":"Domain-wall-synapse-based crossbar arrays have been shown to be very efficient, in terms of speed and energy consumption, while implementing fully connected neural network algorithms for simple data-classification tasks, both in inference and on-chip-learning modes. But for more complex and realistic data-classification tasks, convolutional neural networks (CNN) need to be trained through such crossbar arrays. In this paper, we carry out device–circuit–system co-design and co-simulation of on-chip learning of a CNN using a domain-wall-synapse-based crossbar array. For this purpose, we use a combination of micromagnetic-physics-based synapse-device modeling, SPICE simulation of a crossbar-array circuit using such synapse devices, and system-level-coding using a high-level language. In our design, each synaptic weight of the convolutional kernel is considered to be of 15 bits; one domain-wall-synapse crossbar array is dedicated to the five least significant bits (LSBs), and two crossbar arrays are dedicated to the other bits. The crossbar arrays accelerate the matrix vector multiplication operation involved in the forward computation of the CNN. The synaptic weights of the LSB crossbar are updated after forward computation on every training sample, while the weights of the other crossbars are updated after forward computation on ten samples, to achieve on-chip learning. We report high classification-accuracy numbers for different machine-learning data sets using our method. We also carry out a study of how the classification accuracy of our designed CNN is affected by device-to-device variations, cycle-to-cycle variations, bit precision of the synaptic weights, and the frequency of weight updates.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128385247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Processing IMU action recognition based on brain-inspired computing with microfabricated MEMS resonators","authors":"T. Zheng, Wuhao Yang, Jie Sun, Zhenxi Liu, Kunfeng Wang, X. Zou","doi":"10.1088/2634-4386/ac5ddf","DOIUrl":"https://doi.org/10.1088/2634-4386/ac5ddf","url":null,"abstract":"Reservoir computing (RC) decomposes the recurrent neural network into a fixed network with recursive connections and a trainable linear network. With the advantages of low training cost and easy hardware implementation, it provides a method for the effective processing of time-domain correlation information. In this paper, we build a hardware RC system with a nonlinear MEMS resonator and build an action recognition data set with time-domain correlation. Moreover, two different universal data set are utilized to verify the classification and prediction performance of the RC hardware system. At the same time, the feasibility of the novel data set was validated by three general machine learning approaches. Specifically, the processing of this novel time-domain correlation data set obtained a relatively high success rate. These results, together with the dataset that we build, enable the broad implementation of brain-inspired computing with microfabricated devices, and shed light on the potential for the realization of integrated perception and calculation in our future work.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127035939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect of biologically-motivated energy constraints on liquid state machine dynamics and classification performance","authors":"Andrew Fountain, Cory E. Merkel","doi":"10.1088/2634-4386/ac5d1f","DOIUrl":"https://doi.org/10.1088/2634-4386/ac5d1f","url":null,"abstract":"Equipping edge devices with intelligent behavior opens up new possibilities for automating the decision making in extreme size, weight, and power-constrained application domains. To this end, several recent lines of research are aimed at the design of artificial intelligence hardware accelerators that have significantly reduced footprint and power demands compared to conventional CPU/GPU systems. However, despite some key advancements, the majority of work in this area assumes that there is an unlimited supply of energy available for computation, which is not realistic in the case of battery-powered and energy harvesting devices. In this paper, we address this gap by exploring the computational effects of energy constraints on a popular class of brain-inspired spiking neural networks–liquid state machines (LSMs). Energy constraints were applied by limiting the spiking activity in subsets of LSM neurons. We tested our designs on two biosignal processing tasks: epileptic seizure detection and biometric gait identification. For both tasks, we show that energy constraints can significantly improve classification accuracy. This demonstrates that in the design of neuromorphic systems, reducing energy and increasing performance are not always competing goals.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114306356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect of cycling on ultra-thin HfZrO4, ferroelectric synaptic weights","authors":"L. Bégon-Lours, M. Halter, M. Sousa, Y. Popoff, Diana Dávila Pineda, D. F. Falcone, Zhenming Yu, S. Reidt, Lorenzo Benatti, F. Puglisi, B. Offrein","doi":"10.1088/2634-4386/ac5b2d","DOIUrl":"https://doi.org/10.1088/2634-4386/ac5b2d","url":null,"abstract":"Two-terminal ferroelectric synaptic weights are fabricated on silicon. The active layers consist of a 2 nm thick WO x film and a 2.7 nm thick HfZrO4 (HZO) film grown by atomic layer deposition. The ultra-thin HZO layer is crystallized in the ferroelectric phase using a millisecond flash at a temperature of only 500 °C, evidenced by x-rays diffraction and electron microscopy. The current density is increased by four orders of magnitude compared to weights based on a 5 nm thick HZO film. Potentiation and depression (analog resistive switching) is demonstrated using either pulses of constant duration (as short as 20 nanoseconds) and increasing amplitude, or pulses of constant amplitude (+/−1 V) and increasing duration. The cycle-to-cycle variation is below 1%. Temperature dependent electrical characterisation is performed on a series of device cycled up to 108 times: they reveal that HZO possess semiconducting properties. The fatigue leads to a decrease, in the high resistive state only, of the conductivity and of the activation energy.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125459230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ferroelectric memory based on two-dimensional materials for neuromorphic computing","authors":"Li Chen, M. Pam, Sifan Li, K. Ang","doi":"10.1088/2634-4386/ac57cb","DOIUrl":"https://doi.org/10.1088/2634-4386/ac57cb","url":null,"abstract":"Ferroelectric memory devices with fast-switching speed and ultra-low power consumption have been recognized as promising building blocks for brain-like neuromorphic computing. In particular, ferroelectric memories based on 2D materials are attracting increasing research interest in recent years due to their unique properties that are unattainable in conventional materials. Specifically, the atomically thin 2D materials with tunable electronic properties coupled with the high compatibility with existing complementary metal-oxide-semiconductor technology manifests their potential for extending state-of-the-art ferroelectric memory technology into atomic-thin scale. Besides, the discovery of 2D materials with ferroelectricity shows the potential to realize functional devices with novel structures. This review will highlight the recent progress in ferroelectric memory devices based on 2D materials for neuromorphic computing. The merits of such devices and the range of 2D ferroelectrics being explored to date are reviewed and discussed, which include two- and three-terminal ferroelectric synaptic devices based on 2D materials platform. Finally, current developments and remaining challenges in achieving high-performance 2D ferroelectric synapses are discussed.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122621780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Magnetic tunnel junction based implementation of spike time dependent plasticity learning for pattern recognition","authors":"Aijaz H. Lone, S. Amara, H. Fariborzi","doi":"10.1088/2634-4386/ac57a2","DOIUrl":"https://doi.org/10.1088/2634-4386/ac57a2","url":null,"abstract":"We present a magnetic tunnel junction (MTJ) based implementation of the spike time-dependent (STDP) learning for pattern recognition applications. The proposed hybrid scheme utilizes the spin–orbit torque (SOT) driven neuromorphic device-circuit co-design to demonstrate the Hebbian learning algorithm. The circuit implementation involves the (MTJ) device structure, with the domain wall motion in the free layer, acting as an artificial synapse. The post-spiking neuron behaviour is implemented using a low barrier MTJ. In both synapse and neuron, the switching is driven by the SOTs generated by the spin Hall effect in the heavy metal. A coupled model for the spin transport and switching characteristics in both devices is developed by adopting a modular approach to spintronics. The thermal effects in the synapse and neuron result in a stochastic but tuneable domain wall motion in the synapse and a superparamagnetic behaviour of in neuron MTJ. Using the device model, we study the dimensional parameter dependence of the switching delay and current to optimize the device dimensions. The optimized parameters corresponding to synapse and neuron are considered for the implementation of the Hebbian learning algorithm. Furthermore, cross-point architecture and STDP-based weight modulation scheme is used to demonstrate the pattern recognition capabilities by the proposed neuromorphic circuit.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129015366","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Two-dimensional materials for artificial synapses: toward a practical application","authors":"I-Ting Wang, Chih-Cheng Chang, Yen-Yu Chen, Yi-Shin Su, T. Hou","doi":"10.1088/2634-4386/ac5086","DOIUrl":"https://doi.org/10.1088/2634-4386/ac5086","url":null,"abstract":"Combining the emerging two-dimensional materials (2DMs) and neuromorphic computing, 2DM-based synaptic devices (2DM synapse) are highly anticipated research topics with the promise of revolutionizing the present Si-based computing paradigm. Although the development is still in the early stage, the number of 2DM synapses reported has increased exponentially in the past few years. Nevertheless, most of them mainly focus on device-level synaptic emulations, and a practical perspective toward system-level applications is still lacking. In this review article, we discuss several important types of 2DM synapses for neuromorphic computing. Based on the cross-layer device-circuit-algorithm co-optimization strategy, non-ideal properties in 2DM synapses are considered for accelerating deep neural networks, and their impacts on system-level accuracy, power and area are discussed. Finally, a development guide of 2DM synapses is provided toward accurate online training and inference in the future.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"110 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132323263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Meta-learning spiking neural networks with surrogate gradient descent","authors":"Kenneth Stewart, E. Neftci","doi":"10.1088/2634-4386/ac8828","DOIUrl":"https://doi.org/10.1088/2634-4386/ac8828","url":null,"abstract":"Adaptive ‘life-long’ learning at the edge and during online task performance is an aspirational goal of artificial intelligence research. Neuromorphic hardware implementing spiking neural networks (SNNs) are particularly attractive in this regard, as their real-time, event-based, local computing paradigm makes them suitable for edge implementations and fast learning. However, the long and iterative learning that characterizes state-of-the-art SNN training is incompatible with the physical nature and real-time operation of neuromorphic hardware. Bi-level learning, such as meta-learning is increasingly used in deep learning to overcome these limitations. In this work, we demonstrate gradient-based meta-learning in SNNs using the surrogate gradient method that approximates the spiking threshold function for gradient estimations. Because surrogate gradients can be made twice differentiable, well-established, and effective second-order gradient meta-learning methods such as model agnostic meta learning (MAML) can be used. We show that SNNs meta-trained using MAML perform comparably to conventional artificial neural networks meta-trained with MAML on event-based meta-datasets. Furthermore, we demonstrate the specific advantages that accrue from meta-learning: fast learning without the requirement of high precision weights or gradients, training-to-learn with quantization and mitigating the effects of approximate synaptic plasticity rules. Our results emphasize how meta-learning techniques can become instrumental for deploying neuromorphic learning technologies on real-world problems.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115433395","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Digital coded exposure formation of frames from event-based imagery","authors":"Andy Gothard, Daniel Jones, A. Green, Michael Torrez, A. Cattaneo, D. Mascarenas","doi":"10.1088/2634-4386/ac4917","DOIUrl":"https://doi.org/10.1088/2634-4386/ac4917","url":null,"abstract":"Event-driven neuromorphic imagers have a number of attractive properties including low-power consumption, high dynamic range, the ability to detect fast events, low memory consumption and low band-width requirements. One of the biggest challenges with using event-driven imagery is that the field of event data processing is still embryonic. In contrast, decades worth of effort have been invested in the analysis of frame-based imagery. Hybrid approaches for applying established frame-based analysis techniques to event-driven imagery have been studied since event-driven imagers came into existence. However, the process for forming frames from event-driven imagery has not been studied in detail. This work presents a principled digital coded exposure approach for forming frames from event-driven imagery that is inspired by the physics exploited in a conventional camera featuring a shutter. The technique described in this work provides a fundamental tool for understanding the temporal information content that contributes to the formation of a frame from event-driven imagery data. Event-driven imagery allows for the application of arbitrary virtual digital shutter functions to form the final frame on a pixel-by-pixel basis. The proposed approach allows for the careful control of the spatio-temporal information that is captured in the frame. Furthermore, unlike a conventional physical camera, event-driven imagery can be formed into any variety of possible frames in post-processing after the data is captured. Furthermore, unlike a conventional physical camera, coded-exposure virtual shutter functions can assume arbitrary values including positive, negative, real, and complex values. The coded exposure approach also enables the ability to perform applications of industrial interest such as digital stroboscopy without any additional hardware. The ability to form frames from event-driven imagery in a principled manner opens up new possibilities in the ability to use conventional frame-based image processing techniques on event-driven imagery.","PeriodicalId":198030,"journal":{"name":"Neuromorphic Computing and Engineering","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132527776","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}