Inherent Stochastic Learning in CMOS-Integrated HfO<sub>2</sub> Arrays for Neuromorphic Computing

Wenger, Christian; Zahari, Finn; Mahadevaiah, Mamathamba Kalishettyhalli; Pérez, Eduardo; Darvin, Maxim E.; Kohlstedt, H.; Ziegler, Martin

doi:10.1109/led.2019.2900867

Cited by 28 publications

(26 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address this concern, lowpower neuromorphic platforms could be integrated into medical devices for locally processing of computations required for ML algorithms [32]. Neuromorphic chips have been successfully implemented in different studies for matrixmultiplications required for non-perceptron and perceptronbased ML methods [33], [34]. By bringing the data postprocessing from backend onto a chip, real-time analysis of data in a less time consuming manner with a smaller time delay is feasible.…”

Section: Resultsmentioning

confidence: 99%

In-Vitro Classification of Saliva Samples of COPD Patients and Healthy Controls Using Machine Learning Tools

2020

Self Cite

View full text Add to dashboard Cite

Chronic Obstructive Pulmonary Disease (COPD) is a life-threatening lung disease and a major cause of morbidity and mortality worldwide. Although a curative therapy has yet to be found, permanent monitoring of biomarkers that reflect the disease progression plays a pivotal role for the effective management of COPD. The accurate examination of respiratory tract fluids like saliva is a promising approach for staging disease and predicting its upcoming exacerbations in a Point-of-Care (PoC) environment. However, the concurrent consideration of patients' demographic and medical parameters is necessary for achieving accurate outcomes. Therefore, Machine Learning (ML) tools can play an important role for analyzing patient data and providing comprehensive results for the recognition of COPD in a PoC setting. As a result, the objective of this research work was to implement ML tools on data acquired from characterizing saliva samples of COPD patients and healthy controls as well as their demographic information for PoC recognition of the disease. For this purpose, a permittivity biosensor was used to characterize dielectric properties of saliva samples and, subsequently, ML tools were applied on the acquired data for classification. The XGBoost gradient boosting algorithm provided a high classification accuracy and sensitivity of 91.25% and 100%, respectively, making it a promising model for COPD evaluation. Integration of this model on a neuromorphic chip, in the future, will enable the real-time assessment of COPD in PoC, with low cost, low energy consumption, and high patient privacy. In addition, constant monitoring of COPD in a near-patient setup will enable the better management of the disease exacerbations.

show abstract

Section: Resultsmentioning

confidence: 99%

In-Vitro Classification of Saliva Samples of COPD Patients and Healthy Controls Using Machine Learning Tools

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, this work illustrates promising results for the practicality of using pre-trained neuromorphic chips in complex real-world applications, such as imaging, with time-consuming training requirements. Nevertheless, RRAM neuromorphic systems can also be used for on-chip learning and adaptation to new input patterns by developing network structures and acquiring algorithms 30,37 . Adaptability of these chips to an individual patient data is significantly important for applications such as the epileptic seizure prediction, where developing a generalizable ML model is not possible 38 .…”

Section: Discussionmentioning

confidence: 99%

“…An even better control on the switching event can be achieved by using the Incremental Step Pulse with Verify Algorithm (ISPVA), which was used in this work 45 . It should be noted that applying voltage pulses with lower absolute value of the amplitude leads to stochastic switching between resistance states, which can be exploited for stochastic learning of analog data 30,31 . The stochasticity in amorphous devices is lower than in polycrystalline devices.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

“…A thorough characterization of the devices in terms of switching voltages, endurance, yield and retention is given in 31,43 . For the deployment of the thresholded model with 10 levels, a mixed-signal neuromorphic circuit with software-based neurons and hardware synapses was used similar to those shown in 30 . The RRAM chip was connected via a standard 64 pin integrated circuit socket to a Printed Circuit Board (PCB).…”

Section: Artificial Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Neuromorphic on-chip recognition of saliva samples of COPD and healthy controls using memristive devices

Zarrin

Zahari

Mahadevaiah

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

Chronic Obstructive Pulmonary Disease (COPD) is a life-threatening lung disease, affecting millions of people worldwide. Implementation of Machine Learning (ML) techniques is crucial for the effective management of COPD in home-care environments. However, shortcomings of cloud-based ML tools in terms of data safety and energy efficiency limit their integration with low-power medical devices. To address this, energy efficient neuromorphic platforms can be used for the hardware-based implementation of ML methods. Therefore, a memristive neuromorphic platform is presented in this paper for the on-chip recognition of saliva samples of COPD patients and healthy controls. Results of its performance evaluations showed that the digital neuromorphic chip is capable of recognizing unseen COPD samples with accuracy and sensitivity values of 89% and 86%, respectively. Integration of this technology into personalized healthcare devices will enable the better management of chronic diseases such as COPD.

show abstract

“…Furthermore, RRAM devices are already integrated in CMOS technologies 53 , 54 . While the potential of RRAM cells for stochastic learning was recently shown 47 , a thorough investigation of the concept has yet to be reported. In particular, the influence of the memristive cells’ important technological parameters on their performance in stochastic neural networks has been given very little attention.…”

Section: Introductionmentioning

confidence: 99%

Analogue pattern recognition with stochastic switching binary CMOS-integrated memristive devices

Zahari

Pérez

Mahadevaiah

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Biological neural networks outperform current computer technology in terms of power consumption and computing speed while performing associative tasks, such as pattern recognition. The analogue and massive parallel in-memory computing in biology differs strongly from conventional transistor electronics that rely on the von Neumann architecture. Therefore, novel bio-inspired computing architectures have been attracting a lot of attention in the field of neuromorphic computing. Here, memristive devices, which serve as non-volatile resistive memory, are employed to emulate the plastic behaviour of biological synapses. In particular, CMOS integrated resistive random access memory (RRAM) devices are promising candidates to extend conventional CMOS technology to neuromorphic systems. However, dealing with the inherent stochasticity of resistive switching can be challenging for network performance. In this work, the probabilistic switching is exploited to emulate stochastic plasticity with fully CMOS integrated binary RRAM devices. Two different RRAM technologies with different device variabilities are investigated in detail, and their potential applications in stochastic artificial neural networks (StochANNs) capable of solving MNIST pattern recognition tasks is examined. A mixed-signal implementation with hardware synapses and software neurons combined with numerical simulations shows that the proposed concept of stochastic computing is able to process analogue data with binary memory cells.

show abstract

Inherent Stochastic Learning in CMOS-Integrated HfO₂ Arrays for Neuromorphic Computing

Cited by 28 publications

References 15 publications

In-Vitro Classification of Saliva Samples of COPD Patients and Healthy Controls Using Machine Learning Tools

In-Vitro Classification of Saliva Samples of COPD Patients and Healthy Controls Using Machine Learning Tools

Neuromorphic on-chip recognition of saliva samples of COPD and healthy controls using memristive devices

Analogue pattern recognition with stochastic switching binary CMOS-integrated memristive devices

Contact Info

Product

Resources

About

Inherent Stochastic Learning in CMOS-Integrated HfO2 Arrays for Neuromorphic Computing

Cited by 28 publications

References 15 publications

In-Vitro Classification of Saliva Samples of COPD Patients and Healthy Controls Using Machine Learning Tools

In-Vitro Classification of Saliva Samples of COPD Patients and Healthy Controls Using Machine Learning Tools

Neuromorphic on-chip recognition of saliva samples of COPD and healthy controls using memristive devices

Analogue pattern recognition with stochastic switching binary CMOS-integrated memristive devices

Contact Info

Product

Resources

About

Inherent Stochastic Learning in CMOS-Integrated HfO₂ Arrays for Neuromorphic Computing