GLORIA

GEOMAR Library Ocean Research Information Access

Your email was sent successfully. Check your inbox.

An error occurred while sending the email. Please try again.

Proceed reservation?

Export
Filter
Material
Language
  • 1
    Online Resource
    Online Resource
    IOP Publishing ; 2023
    In:  Chinese Physics B Vol. 32, No. 6 ( 2023-05-01), p. 068704-
    In: Chinese Physics B, IOP Publishing, Vol. 32, No. 6 ( 2023-05-01), p. 068704-
    Abstract: In recent years, spiking neural networks (SNNs) have received increasing attention of research in the field of artificial intelligence due to their high biological plausibility, low energy consumption, and abundant spatio-temporal information. However, the non-differential spike activity makes SNNs more difficult to train in supervised training. Most existing methods focusing on introducing an approximated derivative to replace it, while they are often based on static surrogate functions. In this paper, we propose a progressive surrogate gradient learning for backpropagation of SNNs, which is able to approximate the step function gradually and to reduce information loss. Furthermore, memristor cross arrays are used for speeding up calculation and reducing system energy consumption for their hardware advantage. The proposed algorithm is evaluated on both static and neuromorphic datasets using fully connected and convolutional network architecture, and the experimental results indicate that our approach has a high performance compared with previous research.
    Type of Medium: Online Resource
    ISSN: 1674-1056
    Language: Unknown
    Publisher: IOP Publishing
    Publication Date: 2023
    detail.hit.zdb_id: 2412147-2
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 2
    In: Acta Physica Sinica, Acta Physica Sinica, Chinese Physical Society and Institute of Physics, Chinese Academy of Sciences, Vol. 63, No. 12 ( 2014), p. 128502-
    Abstract: The memristor is a novel kind of electronic device with dynamic variable resistance that is dependent on the past history of the input current or voltage. As the fourth fundamental circuit element, the memristor captures a number of unique properties that have been found to possess attractive potentials in some promising fields such as nonvolatile memory, nonlinear circuit and system, and neuromorphic system. Additionally, compared with a circuit of single memristor, series-parallel circuit of memristors possesses more abundant device characteristics which arouses increasingly extensive interest from numerous researchers. In this paper, the mathematical closed-form charge-governed and flux-governed HP memristor nonlinear models are presented with constructive procedures. In particular, these models are more realistic by taking into account the nonlinear dopant drift effect nearby the terminals and the boundary conditions, and by adding a simple and effective window function. Furthermore, based on the internal parameters and threshold of the memristor respectively, the theoretical derivation and numerical analysis of the memristor-based series-parallel connection circuits have been made comprehensively. For obtaining the characteristics of the memristor-based combinational circuits intuitively, a graphical user interface is designed based on Matlab software, which is beneficial to displaying the properties of the memristive system clearly. The results in the present paper may provide theoretical reference and reliable experimental basis for the further development of the memristor-based combinational circuits.
    Type of Medium: Online Resource
    ISSN: 1000-3290 , 1000-3290
    Language: Unknown
    Publisher: Acta Physica Sinica, Chinese Physical Society and Institute of Physics, Chinese Academy of Sciences
    Publication Date: 2014
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 3
    In: Acta Physica Sinica, Acta Physica Sinica, Chinese Physical Society and Institute of Physics, Chinese Academy of Sciences, Vol. 67, No. 11 ( 2018), p. 110502-
    Abstract: The memristor is a nonlinear element and intrinsically possesses memory function. When it works as nonlinear part of a chaotic system, the complexity and the randomness of signal will be enhanced. In this paper memristor is introduced into a three-dimensional chaotic system based on the augmented L system. The interesting and promising behaviors of complex single, double and four-scroll chaotic attractors generated only by varying a parameter have not been reported in memristive chaotic system and thus they deserve to be further investigated. It is also obvious that such a simple change of one parameter could be used to generate a variety of quite complex attractors. Therefore, as a nonlinear device the memristor plays an important role in this system. Firstly, some basic dynamical properties of the memristive chaotic system, including symmetry and in-variance, the existence of attractor, equilibrium, and stability are investigated in detail. By numerically simulating the power spectrum, Lyapunov exponent, Poincare map and bifurcation diagram, in this paper we verify that the proposed system has abundant dynamical behaviors. The sensitivities of system parameters to the chaotic behaviors are further explored by calculating, in detail, its Lyapunov exponent spectrum and bifurcation diagrams. The results of simulation and experiment are in good agreement, thereby proving the veracity of analysis. The memristive chaotic circuit is designed using the memristor, operational amplifier, analog multiplier and other conventional components. The circuit implementation of the memristive system is simulated using SPICE (simulation program with integrated circuit emphasis). The SPICE simulation results and the theoretical analysis are found to be in good agreement, and thus verifying that the system can produce chaos. Pulse synchronization has the following characteristics: low energy consumption, fast synchronization and easy-to-implement single-channel transmission. Therefore, it is more practical in chaotic secure communication. Subsequently the pulse chaos synchronization is realized from the perspective of the maximum Lyapunov exponent, and numerical simulations show the existence of new memristive chaotic system and the feasibility of pulse synchronization control, and also provide an experimental basis for further studying the applications of the memristive chaotic system in voice secure communication and information processing.
    Type of Medium: Online Resource
    ISSN: 1000-3290 , 1000-3290
    Language: Unknown
    Publisher: Acta Physica Sinica, Chinese Physical Society and Institute of Physics, Chinese Academy of Sciences
    Publication Date: 2018
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 4
    Online Resource
    Online Resource
    China Science Publishing & Media Ltd. ; 2014
    In:  Chinese Journal of Computers Vol. 36, No. 12 ( 2014-3-20), p. 2577-2586
    In: Chinese Journal of Computers, China Science Publishing & Media Ltd., Vol. 36, No. 12 ( 2014-3-20), p. 2577-2586
    Type of Medium: Online Resource
    ISSN: 0254-4164
    Language: English
    Publisher: China Science Publishing & Media Ltd.
    Publication Date: 2014
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 5
    Online Resource
    Online Resource
    IOP Publishing ; 2022
    In:  Chinese Physics B Vol. 31, No. 7 ( 2022-07-01), p. 078702-
    In: Chinese Physics B, IOP Publishing, Vol. 31, No. 7 ( 2022-07-01), p. 078702-
    Abstract: Memristive neural network has attracted tremendous attention since the memristor array can perform parallel multiply-accumulate calculation (MAC) operations and memory-computation operations as compared with digital CMOS hardware systems. However, owing to the variability of the memristor, the implementation of high-precision neural network in memristive computation units is still difficult. Existing learning algorithms for memristive artificial neural network (ANN) is unable to achieve the performance comparable to high-precision by using CMOS-based system. Here, we propose an algorithm based on off-chip learning for memristive ANN in low precision. Training the ANN in the high-precision in digital CPUs and then quantifying the weight of the network to low precision, the quantified weights are mapped to the memristor arrays based on VTEAM model through using the pulse coding weight-mapping rule. In this work, we execute the inference of trained 5-layers convolution neural network on the memristor arrays and achieve an accuracy close to the inference in the case of high precision (64-bit). Compared with other algorithms-based off-chip learning, the algorithm proposed in the present study can easily implement the mapping process and less influence of the device variability. Our result provides an effective approach to implementing the ANN on the memristive hardware platform.
    Type of Medium: Online Resource
    ISSN: 1674-1056
    Language: Unknown
    Publisher: IOP Publishing
    Publication Date: 2022
    detail.hit.zdb_id: 2412147-2
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 6
    Online Resource
    Online Resource
    Elsevier BV ; 2017
    In:  Journal of Physics and Chemistry of Solids Vol. 111 ( 2017-12), p. 41-46
    In: Journal of Physics and Chemistry of Solids, Elsevier BV, Vol. 111 ( 2017-12), p. 41-46
    Type of Medium: Online Resource
    ISSN: 0022-3697
    Language: English
    Publisher: Elsevier BV
    Publication Date: 2017
    detail.hit.zdb_id: 6453-1
    detail.hit.zdb_id: 3114-8
    detail.hit.zdb_id: 1491914-X
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 7
    Online Resource
    Online Resource
    Informa UK Limited ; 2017
    In:  Molecular Physics Vol. 115, No. 20 ( 2017-10-18), p. 2558-2564
    In: Molecular Physics, Informa UK Limited, Vol. 115, No. 20 ( 2017-10-18), p. 2558-2564
    Type of Medium: Online Resource
    ISSN: 0026-8976 , 1362-3028
    Language: English
    Publisher: Informa UK Limited
    Publication Date: 2017
    detail.hit.zdb_id: 1491083-4
    SSG: 12
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 8
    In: Neuropathology, Wiley, Vol. 37, No. 1 ( 2017-02), p. 35-44
    Abstract: Meningeal Ewing Sarcoma (ES)/peripheral primitive neuroectodermal tumor (pPNET) is a rare diagnostically challenging small round cell tumor in the CNS. This study investigates the clinical pathological features of four cases of this tumor from archives of 6 years in our hospital. Patients were within the median age of 21.5 years and male to female ratio was 1:1. The tumors distributed at the supra‐tentorial location, posterior fossa and lumbar vertebral canal, usually presenting as the dura‐sited nodule or having close connection with the meninges within the cranium or vertebral canal. Histopathologically, small round undifferentiated tumor cells with hypercellularities, scant cytoplasm and inconspicuous nucleoli were observed, although some components such as atypical larger vesicular nuclei, prominent nucleoli of tumor cells, necrotic foci and mesenchymal collagen proliferation forming the lobular structure, were also appreciated. Immunohistochemally, tumor cells displayed membranous positivity of CD99 (4/4), nuclear positivity of FLI‐1 (4/4) and NKX2.2 (4/4), negativity of EMA, GFAP and synaptophysin expression. The histochemical PAS staining showed weak positivity in one case. Fluorescence in situ hybridization (FISH) test using EWSR1 (22q12) dual color break apart rearrangement probe showed positive results in two cases. Results suggest that using a panel of immunohistochemical markers, including NKX2.2, CD99, FLI‐1, EMA, GFAP and synaptophysin, combined with the supplementary EWSR1 FISH test, helps to define the diagnosis of meningeal ES/pPNET of CNS.
    Type of Medium: Online Resource
    ISSN: 0919-6544 , 1440-1789
    URL: Issue
    Language: English
    Publisher: Wiley
    Publication Date: 2017
    detail.hit.zdb_id: 2008290-3
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 9
    Online Resource
    Online Resource
    Frontiers Media SA ; 2022
    In:  Frontiers in Plant Science Vol. 13 ( 2022-6-10)
    In: Frontiers in Plant Science, Frontiers Media SA, Vol. 13 ( 2022-6-10)
    Abstract: Deeper and longer roots allow crops to survive and flourish, but our understanding of the plant growth regulators promoting root system establishment is limited. Here, we report that, a novel auxin receptor agonist, named K-10, had a remarkable promotive effect on root growth in both Arabidopsis thaliana and Oryza sativa through the enhancement of root-related signaling responses. Using computer-aided drug discovery approaches, we developed potent lead compound by screening artificial chemicals on the basis of the auxin receptor TIR1 (Transport Inhibitor Response 1), and a series of N-(benzo[d] [1,3] dioxol-5-yl)-2-(one-benzylthio) acetamides, K-1 to K-22, were designed and synthesized. The results of bioassay showed that K-10 exhibited an excellent root growth-promoting activity far exceeding that of NAA (1-naphthylacetic acid). A further morphological investigation of the auxin related mutants ( yucQ , tir1 ) revealed that K-10 had auxin-like physiological functions and was recognized by TIR1, and K-10 significantly enhanced auxin response reporter’s ( DR5:GUS ) transcriptional activity. Consistently, transcriptome analysis showed that K-10 induced a common transcriptional response with auxin and down-regulated the expression of root growth-inhibiting genes. Further molecular docking analysis revealed that K-10 had a stronger binding ability with TIR1 than NAA. These results indicated that this class of derivatives could be a promising scaffold for the discovery and development of novel auxin receptor agonists, and the employment of K-10 may be effective for enhancing root growth and crop production.
    Type of Medium: Online Resource
    ISSN: 1664-462X
    Language: Unknown
    Publisher: Frontiers Media SA
    Publication Date: 2022
    detail.hit.zdb_id: 2687947-5
    detail.hit.zdb_id: 2613694-6
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
  • 10
    Online Resource
    Online Resource
    Elsevier BV ; 2017
    In:  Journal of Alloys and Compounds Vol. 690 ( 2017-01), p. 169-175
    In: Journal of Alloys and Compounds, Elsevier BV, Vol. 690 ( 2017-01), p. 169-175
    Type of Medium: Online Resource
    ISSN: 0925-8388
    Language: English
    Publisher: Elsevier BV
    Publication Date: 2017
    detail.hit.zdb_id: 2012675-X
    Location Call Number Limitation Availability
    BibTip Others were also interested in ...
Close ⊗
This website uses cookies and the analysis tool Matomo. More information can be found here...