Shiv Dutt Joshi

Bishshoy Das, Milton Mondal, Brejesh Lall et al.

In standard neural network training, the gradients in the backward pass are determined by the forward pass. As a result, the two stages are coupled. This is how most neural networks are trained currently. However, gradient modification in the backward pass has seldom been studied in the literature. In this paper we explore decoupled training, where we alter the gradients in the backward pass. We propose a simple yet powerful method called PowerGrad Transform, that alters the gradients before the weight update in the backward pass and significantly enhances the predictive performance of the neural network. PowerGrad Transform trains the network to arrive at a better optima at convergence. It is computationally extremely efficient, virtually adding no additional cost to either memory or compute, but results in improved final accuracies on both the training and test sets. PowerGrad Transform is easy to integrate into existing training routines, requiring just a few lines of code. PowerGrad Transform accelerates training and makes it possible for the network to better fit the training data. With decoupled training, PowerGrad Transform improves baseline accuracies for ResNet-50 by 0.73%, for SE-ResNet-50 by 0.66% and by more than 1.0% for the non-normalized ResNet-18 network on the ImageNet classification task.

Pushpendra Singh, Shiv Dutt Joshi, Rakesh Kumar Patney et al.

Since many decades, there is a general perception in literature that the Fourier methods are not suitable for the analysis of nonlinear and nonstationary data. In this paper, we propose a Fourier Decomposition Method (FDM) and demonstrate its efficacy for the analysis of nonlinear (i.e. data generated by nonlinear systems) and nonstationary time series. The proposed FDM decomposes any data into a small number of `Fourier intrinsic band functions' (FIBFs). The FDM presents a generalized Fourier expansion with variable amplitudes and frequencies of a time series by the Fourier method itself. We propose an idea of zero-phase filter bank based multivariate FDM (MFDM) algorithm, for the analysis of multivariate nonlinear and nonstationary time series, from the FDM. We also present an algorithm to obtain cutoff frequencies for MFDM. The MFDM algorithm is generating finite number of band limited multivariate FIBFs (MFIBFs). The MFDM preserves some intrinsic physical properties of the multivariate data, such as scale alignment, trend and instantaneous frequency. The proposed methods produce the results in a time-frequency-energy distribution that reveal the intrinsic structures of a data. Simulations have been carried out and comparison is made with the Empirical Mode Decomposition (EMD) methods in the analysis of various simulated as well as real life time series, and results show that the proposed methods are powerful tools for analyzing and obtaining the time-frequency-energy representation of any data.

Pushpendra Singh, Shiv Dutt Joshi

In this paper, we propose the Fourier frequency vector (FFV), inherently, associated with multidimensional Fourier transform. With the help of FFV, we are able to provide physical meaning of so called negative frequencies in multidimensional Fourier transform (MDFT), which in turn provide multidimensional spatial and space-time series analysis. The complex exponential representation of sinusoidal function always yields two frequencies, negative frequency corresponding to positive frequency and vice versa, in the multidimensional Fourier spectrum. Thus, using the MDFT, we propose multidimensional Hilbert transform (MDHT) and associated multidimensional analytic signal (MDAS) with following properties: (a) the extra and redundant positive, negative, or both frequencies, introduced due to complex exponential representation of multidimensional Fourier spectrum, are suppressed, (b) real part of MDAS is original signal, (c) real and imaginary part of MDAS are orthogonal, and (d) the magnitude envelope of a original signal is obtained as the magnitude of its associated MDAS, which is the instantaneous amplitude of the MDAS. The proposed MDHT and associated DMAS are generalization of the 1D HT and AS, respectively. We also provide the decomposition of an image into the AM-FM image model by the Fourier method and obtain explicit expression for the analytic image computation by 2DDFT.

Pushpendra Singh, Shiv Dutt Joshi, Rakesh Kumar Patney et al.

In this paper, we propose algorithms which preserve energy in empirical mode decomposition (EMD), generating finite $n$ number of band limited Intrinsic Mode Functions (IMFs). In the first energy preserving EMD (EPEMD) algorithm, a signal is decomposed into linearly independent (LI), non orthogonal yet energy preserving (LINOEP) IMFs and residue (EPIMFs). It is shown that a vector in an inner product space can be represented as a sum of LI and non orthogonal vectors in such a way that Parseval's type property is satisfied. From the set of $n$ IMFs, through Gram-Schmidt orthogonalization method (GSOM), $n!$ set of orthogonal functions can be obtained. In the second algorithm, we show that if the orthogonalization process proceeds from lowest frequency IMF to highest frequency IMF, then the GSOM yields functions which preserve the properties of IMFs and the energy of a signal. With the Hilbert transform, these IMFs yield instantaneous frequencies and amplitudes as functions of time that reveal the imbedded structures of a signal. The instantaneous frequencies and square of amplitudes as functions of time produce a time-frequency-energy distribution, referred as the Hilbert spectrum, of a signal. Simulations have been carried out for the analysis of various time series and real life signals to show comparison among IMFs produced by EMD, EPEMD, ensemble EMD and multivariate EMD algorithms. Simulation results demonstrate the power of this proposed method.