SPEAKER IDENTIFICATION MODEL BASED ON DEEP NURAL NETWOKS

This study aims is to establish a small system of text-independent recognition of speakers for a relatively small group of speakers at a sound stage. The fascinating justification for the International Space Station (ISS) to detect if the astronauts are speaking at a specific time has influenced the difficulty. In this work, we employed Machine Learning Applications. Accordingly, we used the Direct Deep Neural Network (DNN)-based approach, in which the posterior opportunities of the output layer are utilized to determine the speaker’s presence. In line with the small footprint design objective, a simple DNN model with only sufficient hidden units or sufficient hidden units per layer was designed, thereby reducing the cost of parameters through intentional preparation to avoid the normal overfitting problem and optimize the algorithmic aspects, such as context-based training, activation functions, validation, and learning rate. Two commercially available databases, namely, TIMIT clean speech and HTIMIT multihandset communication database and TIMIT noise-added data framework, were tested for this reference model that we developed using four sound categories at three distinct signal-to-noise ratios. Briefly, we used a dynamic pruning method in which the conditions of all layers are simultaneously pruned, and the pruning mechanism is reassigned. The usefulness of this approach was evaluated on all the above contact databases

Interval universal approximation for neural networks

To verify safety and robustness of neural networks, researchers have successfully applied abstract interpretation , primarily using the interval abstract domain. In this paper, we study the theoretical power and limits of the interval domain for neural-network verification. First, we introduce the interval universal approximation (IUA) theorem. IUA shows that neural networks not only can approximate any continuous function f (universal approximation) as we have known for decades, but we can find a neural network, using any well-behaved activation function, whose interval bounds are an arbitrarily close approximation of the set semantics of f (the result of applying f to a set of inputs). We call this notion of approximation interval approximation . Our theorem generalizes the recent result of Baader et al. from ReLUs to a rich class of activation functions that we call squashable functions . Additionally, the IUA theorem implies that we can always construct provably robust neural networks under ℓ ∞ -norm using almost any practical activation function. Second, we study the computational complexity of constructing neural networks that are amenable to precise interval analysis. This is a crucial question, as our constructive proof of IUA is exponential in the size of the approximation domain. We boil this question down to the problem of approximating the range of a neural network with squashable activation functions. We show that the range approximation problem (RA) is a Δ 2 -intermediate problem, which is strictly harder than NP -complete problems, assuming coNP ⊄ NP . As a result, IUA is an inherently hard problem : No matter what abstract domain or computational tools we consider to achieve interval approximation, there is no efficient construction of such a universal approximator. This implies that it is hard to construct a provably robust network, even if we have a robust network to start with.

Multimodal transistors as ReLU activation functions in physical neural network classifiers

Artificial neural networks (ANNs) providing sophisticated, power-efficient classification are finding their way into thin-film electronics. Thin-film technologies require robust, layout-efficient devices with facile manufacturability. Here, we show how the multimodal transistor’s (MMT’s) transfer characteristic, with linear dependence in saturation, replicates the rectified linear unit (ReLU) activation function of convolutional ANNs (CNNs). Using MATLAB, we evaluate CNN performance using systematically distorted ReLU functions, then substitute measured and simulated MMT transfer characteristics as proxies for ReLU. High classification accuracy is maintained, despite large variations in geometrical and electrical parameters, as CNNs use the same activation functions for training and classification.

Functioning Fuzzy Logic in Optimizing the Solar Systems Work

Fuzzy logic has been used in many fields, either to control a specific movement, improve the productivity of a machine, or monitor the work of an electrical or mechanical system or the like. In this chapter, we will discuss what are the basic factors that must be taken to use the fuzzy logic in the aforementioned matters in general, and then focus on its employment in the field of renewable energy. Three main axes for renewable energy are solar panels, a wind turbine and finally, solar collectors. The key to working and the basis of the static system is the mechanism for selecting the inputs that directly affect the output in addition to the methods and activation functions of the fuzzy logic.

A novel ensemble method based on SBLMD-ANN-MOPSO approach for predicting the milling stability regimes

In recent decades, lots of work has been done to mitigate self excited vibration effects in milling operations. Still, a robust methodology is yet to be developed that can suggest stability bounds pertaining to higher metal removal rate (MRR). In the present work, experimentally acquired acoustic signals in milling operation have been computed using a modified Local Mean Decomposition (SBLMD) technique in order to cite tool chatter features. Further, three artificial neural network (ANN) training algorithms viz. Resilient Propagation (RP), Conjugate Gradient-Based (CGP) and Levenberg-Marquardt Algorithm (LM) and two activation functions viz. Hyperbolic Tangent Sigmoid (TANSIG) and Log Sigmoid (LOGSIG) has been used to train the acquired chatter vibration and metal removal rate data set. Over-fitting or under-fitting issues may arise from the random selection of a number of hidden neurons. The solution to these problems is also proposed in this paper. Among these training algorithms and activation functions, a suitable one has been selected and further invoked to develop prediction models of chatter severity and metal removal rate. Finally, Multi-Objective Particle Swarm Optimization (MOPSO) has been invoked to optimize developed prediction models for obtaining the most favourable range of input parameters pertaining to stable milling with higher productivity.

Cryptocurrency Price Estimation Using Hyperparameterized Oscillatory Activation Functions in LSTM Networks

Activation functions are critical components of neural networks, helping the model learn highly-intricate dependencies, trends, and patterns. Non-linear activation functions allow the model to behave as a functional approximator, learning complex decision boundaries and multi-dimensional patterns in the data. Activation functions can be combined with one another to learn better representations with the objective of improving gradient flow, performance metrics reducing training time and computational cost. Recent work on oscillatory activation functions\cite{noel2021growing}\cite{noel2021biologically} showcased their ability to perform competitively on image classification tasks using a compact architecture. Our work proposes the utilization of these oscillatory activation functions for predicting the volume-weighted average of Bitcoin on the G-Research Cryptocurrency Dataset. We utilize a popular LSTM architecture for this task achieving competitive results when compared to popular activation functions formally used.