Приказано за аутора "Savić, Milan"

Приказ резултата 1-7 од 7

    • Binary Quantization Analysis of Neural Networks Weights on MNIST Dataset 

      Perić, Zoran; Denić, Bojan; Savić, Milan; Vučić, Nikola; Simić, Nikola (Kaunas University of Technology, 2021-08-23)
      This paper considers the design of a binary scalar quantizer of Laplacian source and its application in compressed neural networks. The quantizer performance is investigated in a wide dynamic range of data variances, and ...

    • Design and Analysis of Binary Scalar Quantizer of Laplacian Source with Applications 

      Perić, Zoran; Denić, Bojan; Savić, Milan; Despotović, Vladimir (MDPI, 2020-10-27)
      A compression method based on non-uniform binary scalar quantization, designed for the memoryless Laplacian source with zero-mean and unit variance, is analyzed in this paper. Two quantizer design approaches are presented ...

    • Design of a 2-Bit Neural Network Quantizer for Laplacian Source 

      Perić, Zoran; Savić, Milan; Simić, Nikola; Denić, Bojan; Despotović, Vladimir (Molecular Diversity Preservation International, 2021-07-22)
      Achieving real-time inference is one of the major issues in contemporary neural network applications, as complex algorithms are frequently being deployed to mobile devices that have constrained storage and computing power. ...

    • Influence of zero boresight and non-zero boresight pointing error on BER FSO transmission over Malaga atmospheric channel 

      Smilić, Marko; Perić, Zoran; Milić, Dejan; Marković, Aleksandar; Savić, Milan (Oficyna Wydawnicza Politechniki Wroclawskiej, 2018-03-22)
      In this paper, the influence of a zero and a non-zero boresight pointing errors on the performances of free-space optic transmission over the Málaga atmospheric turbulence channel is considered. Closed form expressions for ...

    • Quantization of Weights of Neural Networks with Negligible Decreasing of Prediction Accuracy 

      Perić, Zoran; Denić, Bojan; Savić, Milan; Dinčić, Milan; Mihajlov, Darko (Kaunas University of Technology, 2021-09-24)
      Quantization and compression of neural network parameters using the uniform scalar quantization is carried out in this paper. The attractiveness of the uniform scalar quantizer is reflected in a low complexity and relatively ...

    • Speaker Recognition Using Constrained Convolutional Neural Networks in Emotional Speech 

      Simić, Nikola; Suzić, Siniša; Nosek, Tijana; Vujović, Mia; Perić, Zoran; Savić, Milan; Delić, Vlado (Molecular Diversity Preservation International, 2022-03-16)
      Speaker recognition is an important classification task, which can be solved using several approaches. Although building a speaker recognition model on a closed set of speakers under neutral speaking conditions is a ...

    • SSC Reception over Kappa-Mu Shadowed Fading Channels in the Presence of Multiple Rayleigh Interferers 

      Stevanović, Aco; Panić, Stefan; Spalević, Petar; Prlinčević, Bojan; Savić, Milan (Kaunas University of Technology, 2018-06-22)
      Modelling of complex wireless transmission scenario over fading channels in the presence of multiple co- channel (CCI) interferers will be presented in this paper. Propagation environment will be modelled with general ...