The Experts below are selected from a list of 7557 Experts worldwide ranked by ideXlab platform

Shimeng Yu - One of the best experts on this subject based on the ideXlab platform.

  • Weight Tuning of Resistive Synaptic Devices and Convolution Kernel Operation on 12 × 12 Cross-Point Array
    Neuro-inspired Computing Using Resistive Synaptic Devices, 2017
    Co-Authors: Shimeng Yu
    Abstract:

    Analog conductance of resistive random access memory (RRAM) is attractive for implementing the synaptic weights in neuro-inspired algorithms. One of the most popular deep learning algorithms is the Convolutional neural network (CNN). The implementation of the Convolution Kernel on the resistive cross-point array is different than the implementation of the matrix-vector multiplication in prior works. In this chapter, we review our recent progress on offline weight tuning of the RRAM and its application for Convolution Kernel operation. First, we developed an optimized iterative programming protocol to tune the weights of HfOx-based RRAM by adjusting the pulse amplitude incremental steps, the pulse width incremental steps, and the start voltages. Then, we demonstrated the key operation in the CNN—the Convolution Kernel on a 12 × 12 cross-point array. We proposed a dimensional reduction of 2D Kernel matrix into 1D column vector, i.e., a column of the array, and enable the parallel readout of multiple 2D Kernels simultaneously. As a proof-of-concept demonstration, we used the offline trained edge filters to detect both horizontal and vertical edges of the 20 × 20 pixels of black-and-white MNIST handwritten digits and the 50 × 50 pixels of a grayscale “dog” image on a 12 × 12 resistive cross-point array based on the HfOx RRAM. The experimental results of the Kernel operation perfectly match the simulation results, indicating the feasibility of the proposed implementation methodology of the Convolution Kernel on the resistive cross-point array for future large-scale integration.

  • demonstration of Convolution Kernel operation on resistive cross point array
    IEEE Electron Device Letters, 2016
    Co-Authors: Pai-yu Chen, Shimeng Yu
    Abstract:

    Convolution is the key operation in the Convolutional neural network, one of the most popular deep learning algorithms. The implementation of the Convolution Kernel on the resistive cross-point array is different than the implementation of the matrix-vector multiplication in prior works. In this letter, we propose a dimensional reduction of 2-D Kernel matrix into 1-D column vector, i.e., a column of the array, and enable the parallel readout of multiple 2-D Kernels simultaneously. As a proof-of-concept demonstration, we use the Prewitt Kernels to detect both horizontal and vertical edges of the $20 \times 20$ pixels of black-and-white MNIST handwritten digits. The experiments were performed on the fabricated $12 \times 12$ resistive cross-point array based on the Pt/HfO x /TiN structure. The experimental results of the Prewitt Kernel operation perfectly matches the simulation results, indicating the feasibility of the proposed implementation methodology of the Convolution Kernel on resistive cross-point array.

  • Demonstration of Convolution Kernel Operation on Resistive Cross-Point Array
    IEEE Electron Device Letters, 2016
    Co-Authors: Pai-yu Chen, Shimeng Yu
    Abstract:

    Convolution is the key operation in the Convolutional neural network, one of the most popular deep learning algorithms. The implementation of the Convolution Kernel on the resistive cross-point array is different than the implementation of the matrix-vector multiplication in prior works. In this letter, we propose a dimensional reduction of 2-D Kernel matrix into 1-D column vector, i.e., a column of the array, and enable the parallel readout of multiple 2-D Kernels simultaneously. As a proof-of-concept demonstration, we use the Prewitt Kernels to detect both horizontal and vertical edges of the 20 × 20 pixels of black and-white MNIST handwritten digits. The experiments were performed on the fabricated 12 × 12 resistive cross-point array based on the Pt/HfOx/TiN structure. The experimental results of the Prewitt Kernel operation perfectly matches the simulation results, indicating the feasibility of the proposed implementation methodology of the Convolution Kernel on resistive cross-point array.

  • Weight tuning of resistive memories and Convolution Kernel operation on cross-point array for neuro-inspired computing
    2016 13th IEEE International Conference on Solid-State and Integrated Circuit Technology (ICSICT), 2016
    Co-Authors: Pai-yu Chen, Shimeng Yu
    Abstract:

    Analog conductance of resistive memories is attractive for implementing the weights in neuro-inspired algorithms. One of the most popular deep learning algorithms is the Convolutional neural network (CNN). In this paper, we review our recent progress on using resistive memories for neuro-inspired computing. First, we optimized the iterative programming protocol to tune the weights of HfOx based resistive memories by adjusting the pulse amplitude incremental steps, the pulse width incremental steps, and the start voltages. Then, we demonstrated the key operation in the CNN - the Convolution Kernel on a 12×12 cross-point array. As a proof-of-concept demonstration, we use the offline trained edge filters to detect both horizontal and vertical edges of the 50×50 pixels of a grayscale “dog” image. The experimental Kernel operation matches the simulation results.

Pai-yu Chen - One of the best experts on this subject based on the ideXlab platform.

  • demonstration of Convolution Kernel operation on resistive cross point array
    IEEE Electron Device Letters, 2016
    Co-Authors: Pai-yu Chen, Shimeng Yu
    Abstract:

    Convolution is the key operation in the Convolutional neural network, one of the most popular deep learning algorithms. The implementation of the Convolution Kernel on the resistive cross-point array is different than the implementation of the matrix-vector multiplication in prior works. In this letter, we propose a dimensional reduction of 2-D Kernel matrix into 1-D column vector, i.e., a column of the array, and enable the parallel readout of multiple 2-D Kernels simultaneously. As a proof-of-concept demonstration, we use the Prewitt Kernels to detect both horizontal and vertical edges of the $20 \times 20$ pixels of black-and-white MNIST handwritten digits. The experiments were performed on the fabricated $12 \times 12$ resistive cross-point array based on the Pt/HfO x /TiN structure. The experimental results of the Prewitt Kernel operation perfectly matches the simulation results, indicating the feasibility of the proposed implementation methodology of the Convolution Kernel on resistive cross-point array.

  • Demonstration of Convolution Kernel Operation on Resistive Cross-Point Array
    IEEE Electron Device Letters, 2016
    Co-Authors: Pai-yu Chen, Shimeng Yu
    Abstract:

    Convolution is the key operation in the Convolutional neural network, one of the most popular deep learning algorithms. The implementation of the Convolution Kernel on the resistive cross-point array is different than the implementation of the matrix-vector multiplication in prior works. In this letter, we propose a dimensional reduction of 2-D Kernel matrix into 1-D column vector, i.e., a column of the array, and enable the parallel readout of multiple 2-D Kernels simultaneously. As a proof-of-concept demonstration, we use the Prewitt Kernels to detect both horizontal and vertical edges of the 20 × 20 pixels of black and-white MNIST handwritten digits. The experiments were performed on the fabricated 12 × 12 resistive cross-point array based on the Pt/HfOx/TiN structure. The experimental results of the Prewitt Kernel operation perfectly matches the simulation results, indicating the feasibility of the proposed implementation methodology of the Convolution Kernel on resistive cross-point array.

  • Weight tuning of resistive memories and Convolution Kernel operation on cross-point array for neuro-inspired computing
    2016 13th IEEE International Conference on Solid-State and Integrated Circuit Technology (ICSICT), 2016
    Co-Authors: Pai-yu Chen, Shimeng Yu
    Abstract:

    Analog conductance of resistive memories is attractive for implementing the weights in neuro-inspired algorithms. One of the most popular deep learning algorithms is the Convolutional neural network (CNN). In this paper, we review our recent progress on using resistive memories for neuro-inspired computing. First, we optimized the iterative programming protocol to tune the weights of HfOx based resistive memories by adjusting the pulse amplitude incremental steps, the pulse width incremental steps, and the start voltages. Then, we demonstrated the key operation in the CNN - the Convolution Kernel on a 12×12 cross-point array. As a proof-of-concept demonstration, we use the offline trained edge filters to detect both horizontal and vertical edges of the 50×50 pixels of a grayscale “dog” image. The experimental Kernel operation matches the simulation results.

A. Holobar - One of the best experts on this subject based on the ideXlab platform.

  • non invasive decoding of the motoneurons a guided source separation method based on Convolution Kernel compensation with clustered initial points
    Frontiers in Computational Neuroscience, 2019
    Co-Authors: Mohammad Reza Mohebian, Hamid Reza Marateb, Saeed Karimimehr, Miquel Angel Mananas, Jernej Kranjec, A. Holobar
    Abstract:

    Despite the progress in understanding of neural codes, the studies of the cortico-muscular coupling still largely rely on interferential electromyographic (EMG) signal or its rectification for the assessment of motor neuron pool behavior. This assessment is non-trivial and should be used with precaution. Direct analysis of neural codes by decomposing the EMG, also known as neural decoding, is an alternative to EMG amplitude estimation. In this study, we propose a fully-deterministic hybrid surface EMG (sEMG) decomposition approach that combines the advantages of both template-based and Blind Source Separation (BSS) decomposition approaches, a.k.a. guided source separation (GSS), to identify motor unit (MU) firing patterns. We use the single-pass density-based clustering algorithm to identify possible cluster representatives in different sEMG channels. These cluster representatives are then used as initial points of modified gradient Convolution Kernel Compensation (gCKC) algorithm. Afterwards, we use the Kalman filter to reduce the noise impact and increase convergence rate of MU filter identification by gCKC. Moreover, we designed an adaptive soft-thresholding method to identify MU firing times out of estimated MU spike trains. We tested the proposed algorithm on a set of synthetic sEMG signals with known MU firing patterns. A grid of 9×10 monopolar surface electrodes with 5-mm inter-electrode distances in both directions was simulated. Muscle excitation was set to 10%, 30% and 50%. Colored Gaussian zero-mean noise with the signal-to-noise ratio (SNR) of 10 dB, 20 dB and 30 dB, respectively, was added to 16 s long sEMG signals that were sampled at 4096 Hz. Overall, 45 simulated signals were analyzed. Our decomposition approach was compared with gCKC algorithm. Overall, in our algorithm, the average numbers of identified MUs and Rate-of-Agreement (RoA) were 16.41 ± 4.18 MUs and 84.00 ± 0.06 %, respectively, whereas the gCKC identified 12.10 ± 2.32 MUs with the average RoA of 90.78±0.08 %. Therefore, the proposed GSS method identified more MUs than the gCKC, with comparable performance. Its performance was dependent on the signal quality but not the signal complexity at different force levels. The proposed algorithm is a promising new offline tool in clinical neurophysiology.

  • progressive fastica peel off and Convolution Kernel compensation demonstrate high agreement for high density surface emg decomposition
    Neural Plasticity, 2016
    Co-Authors: Maoqi Chen, A. Holobar, Xu Zhang, Ping Zhou
    Abstract:

    Decomposition of electromyograms (EMG) is a key approach to investigating motor unit plasticity. Various signal processing techniques have been developed for high density surface EMG decomposition, among which the Convolution Kernel compensation (CKC) has achieved high decomposition yield with extensive validation. Very recently, a progressive FastICA peel-off (PFP) framework has also been developed for high density surface EMG decomposition. In this study, the CKC and PFP methods were independently applied to decompose the same sets of high density surface EMG signals. Across 91 trials of 64-channel surface EMG signals recorded from the first dorsal interosseous (FDI) muscle of 9 neurologically intact subjects, there were a total of 1477 motor units identified from the two methods, including 969 common motor units. On average, common motor units were identified from each trial, which showed a very high matching rate of % in their discharge instants. The high degree of agreement of common motor units from the CKC and the PFP processing provides supportive evidence of the decomposition accuracy for both methods. The different motor units obtained from each method also suggest that combination of the two methods may have the potential to further increase the decomposition yield.

  • gradient Convolution Kernel compensation applied to surface electromyograms
    International Conference on Independent Component Analysis and Signal Separation, 2007
    Co-Authors: A. Holobar, D. Zazula
    Abstract:

    This paper introduces gradient based method for robust assessment of the sparse pulse sources, such as motor unit innervation pulse trains in the filed of electromyography. The method employs multichannel recordings and is based on Convolution Kernel Compensation (CKC). In the first step, the unknown mixing channels (Convolution Kernels) are compensated, while in the second step the natural gradient algorithm is used to blindly optimize the estimated source pulse trains. The method was tested on the simulated mixtures with random mixing matrices, on synthetic surface electromyograms and on real surface electromyograms, recorded from the external anal sphincter muscle. The results prove the method is highly robust to noise and enables complete reconstruction of up to 10 concurrently active motor units.

  • ICA - Gradient Convolution Kernel compensation applied to surface electromyograms
    Independent Component Analysis and Signal Separation, 2007
    Co-Authors: A. Holobar, D. Zazula
    Abstract:

    This paper introduces gradient based method for robust assessment of the sparse pulse sources, such as motor unit innervation pulse trains in the filed of electromyography. The method employs multichannel recordings and is based on Convolution Kernel Compensation (CKC). In the first step, the unknown mixing channels (Convolution Kernels) are compensated, while in the second step the natural gradient algorithm is used to blindly optimize the estimated source pulse trains. The method was tested on the simulated mixtures with random mixing matrices, on synthetic surface electromyograms and on real surface electromyograms, recorded from the external anal sphincter muscle. The results prove the method is highly robust to noise and enables complete reconstruction of up to 10 concurrently active motor units.

  • multichannel blind source separation using Convolution Kernel compensation
    IEEE Transactions on Signal Processing, 2007
    Co-Authors: A. Holobar, D. Zazula
    Abstract:

    This paper studies a novel decomposition technique, suitable for blind separation of linear mixtures of signals comprising finite-length symbols. The observed symbols are first modeled as channel responses in a multiple-input-multiple-output (MIMO) model, while the channel inputs are conceptually considered sparse positive pulse trains carrying the information about the symbol arising times. Our decomposition approach compensates channel responses and aims at reconstructing the input pulse trains directly. The algorithm is derived first for the overdetermined noiseless MIMO case. A generalized scheme is then provided for the underdetermined mixtures in noisy environments. Although blind, the proposed technique approaches Bayesian optimal linear minimum mean square error estimator and is, hence, significantly noise resistant. The results of simulation tests prove it can be applied to considerably underdetermined convolutive mixtures and even to the mixtures of moderately correlated input pulse trains, with their cross-correlation up to 10% of its maximum possible value.

Marián Slodička - One of the best experts on this subject based on the ideXlab platform.

Ping Zhou - One of the best experts on this subject based on the ideXlab platform.

  • progressive fastica peel off and Convolution Kernel compensation demonstrate high agreement for high density surface emg decomposition
    Neural Plasticity, 2016
    Co-Authors: Maoqi Chen, A. Holobar, Xu Zhang, Ping Zhou
    Abstract:

    Decomposition of electromyograms (EMG) is a key approach to investigating motor unit plasticity. Various signal processing techniques have been developed for high density surface EMG decomposition, among which the Convolution Kernel compensation (CKC) has achieved high decomposition yield with extensive validation. Very recently, a progressive FastICA peel-off (PFP) framework has also been developed for high density surface EMG decomposition. In this study, the CKC and PFP methods were independently applied to decompose the same sets of high density surface EMG signals. Across 91 trials of 64-channel surface EMG signals recorded from the first dorsal interosseous (FDI) muscle of 9 neurologically intact subjects, there were a total of 1477 motor units identified from the two methods, including 969 common motor units. On average, common motor units were identified from each trial, which showed a very high matching rate of % in their discharge instants. The high degree of agreement of common motor units from the CKC and the PFP processing provides supportive evidence of the decomposition accuracy for both methods. The different motor units obtained from each method also suggest that combination of the two methods may have the potential to further increase the decomposition yield.