Neural network Abstract A neural network is a data processing system made up of a large number of simple, highly interconnected processing components arranged in an architecture inspired by the cerebral cortex. As a result, neural networks are often capable of performing tasks that humans or animals excel at but that traditional computers struggle with. In recent years, neural networks have risen to prominence as a promising area for research, development, and application to a wide range of real-world situations. Indeed, neural networks have qualities and capacities that no other technology can match. Reading characters and human handwriting, reading typewritten text, compensating for robot alignment errors, interpreting very "noisy" signals (e.g. electrocardiograms), modeling complex systems that cannot be modelled mathematically, and predicting whether proposed loans will be good or fail are just a few examples. This paper provides a basic introduction to neural networks and outlines a few applications. Introduction Neural networks are set of algorithms inspired by the functioning of human brian used in Machine Learning. Generally when you open your eyes, what you see is called data and is processed by the Nuerons(data processing cells) in your brain, and recognises what is around you. That’s how similar the Neural Networks works. They takes a large set of data, process the data (draws out the patterns from data), and outputs what it is. Keywords Neuron: A building block of ANN. It is responsible for accepting input data, performing calculations, and producing output. Input data: Information or data provided to the neurons. Artificial Neural Network(ANN): A computational system inspired by the way biological neural networks in the human brain process information. Deep Neural Network: An ANN with many layers placed between the input layer and the output layer. Weights: The strength of the connection between two neurons. Weights determine what impact the input will have on the output. Bias: An additional parameter used along with the sum of the product of weights and inputs to produce an output. Activation Function: Determines the output of a neural network. Convolutional Neural Networks (CNN): class of artificial neural network, most commonly applied to analyze visual imagery. Multilayer Perceptron (MLP): feedforward artificial neural network that generates a set of outputs from a set of inputs and is characterized by several layers of input nodes connected as a directed graph between the input and output layers. Recurrent Neural Networks (RNN): artificial neural network which uses sequential data or time series data. What are Neural networks? A neural network is a network or circuit of neurons, or in today's terms, an artificial neural network made up of artificial neurons or nodes. Thus, a neural network can be either biological (made up of biological neurons) or artificial (made up of artificial neurons) and used to solve artificial intelligence (AI) challenges. Artificial neural networks (ANNs), usually simply called neural networks (NNs), are computing systems inspired by the biological neural networks that constitute animal brains. An ANN is based on a collection of connected units or nodes called artificial neurons, which loosely model the neurons in a biological brain. Each link can send a signal to other neurons, just like synapses in a human brain. An artificial neuron receives a signal, analyses it, and then sends signals to neurons it is connected to. Each neuron's output is generated by some non-linear function of the sum of its inputs, and the "signal" at a connection is a real number. Edges are the terms for the connections. The weight of neurons and edges is frequently adjusted as learning progresses. The signal strength at a connection is increased or decreased by the weight. Neurons may have a threshold that allows them to send a signal only if the aggregate signal exceeds it. Neurons are usually grouped into layers. On their inputs, separate layers may apply different transformations. After traversing the layers numerous times, you should arrive at the last layer (the output layer). A biological neural network is made up of a collection of chemically or functionally related neurons. A single neuron may be connected to a large number of other neurons, and a network's total number of neurons and connections may be large. Synapses are produced when axons attach to dendrites, while dendrodendritic synapses and other interactions are also conceivable. Other types of signaling emerge from neurotransmitter diffusion in addition to electrical signaling. Artificial intelligence, cognitive modeling, and neural networks are data-processing paradigms inspired by biological neural systems. Artificial intelligence and cognitive modeling attempt to replicate some of the characteristics of biological neural networks. Artificial neural networks have been effectively applied to speech recognition, image analysis, and adaptive control in the artificial intelligence area to create software agents (in computer and video games) or autonomous robots. History of Neural Networks Warren McCulloch, a neurophysiologist, and Walter Pitts, a young mathematician, published a paper on how neurons could work in 1943, which was the first step toward neural networks. They used electrical circuits to model a rudimentary neural network. In his book The Organization of Behavior, published in 1949, Donald Hebb reaffirmed the concept of neurons. It was pointed out that each time brain pathways are used, they are reinforced. Nathanial Rochester of IBM's research laboratories conducted the first attempt to simulate a neural network in the 1950s. Dartmouth's Summer Research Project on Artificial Intelligence, which began in 1956, gave artificial intelligence and neural networks a boost. This sparked interest in AI and the brain's lower-level neural processing system. In 1957, John von Neumann proposed utilizing telegraph relays or vacuum tubes to mimic simple neuron operations. Frank Rosenblatt, a Cornell neurobiologist, began working on the Perceptron in 1958. The operation of a fly's eye piqued his interest. A fly's eye does a lot of the processing that signals it to flee. The Perceptron, which was developed as a result of this study, is the oldest neural network still in use today. A single-layer perceptron was discovered to be effective at classifying a continuous-valued set of inputs into one of two categories. The perceptron calculates a weighted total of the inputs, subtracts a threshold, and then outputs one of two possible results. Fear, unfulfilled promises, and other factors hindered neural network research until 1981. John Hopfield then presented a paper to the National Academy of Sciences in 1982. He was pleasant, articulate, and charismatic in his approach to creating helpful technologies. When Japan unveiled their Fifth-Generation effort at the US-Japan Joint Conference on Cooperative/Competitive Neural Networks in 1982, the US was concerned about falling behind. Funding was soon coming again. The American Institute of Physics started a symposium called Neural Networks for Computing in 1985, which has since become an annual event. The first International Conference on Neural Networks, held by the Institute of Electrical and Electronic Engineers (IEEE) in 1987, gathered over 1,800 people. Schmidhuber & Hochreiter proposed Long Short-Term Memory (LSTM) as a recurrent neural network framework in 1997. In the modern era neural networks are assisting humans to survive the new age transitions in education, financial, aerospace and automotive sectors. What are Neurons in a Neural Network? A layer is made up of microscopic units known as neurons. With the help of biological neurons, a neuron in a neural network can be better comprehended. A biological neuron is analogous to an artificial neuron. It takes in information from other neurons, processes it, and then provides an output. Here, X1 and X2 are inputs to the artificial neurons, f(X) represents the processing done on the inputs and y represents the output of the neuron. How a Neural Network Works? There are several layers in a neural network. Each layer serves a distinct purpose, and the more levels there are, the more complex the network becomes. A neural network is also known as a multi-layer perceptron because of this. A neural network in its purest form includes three layers: 1. The input layer 2. The hidden layer 3. The output layer Each of these levels serves a specific role, as their names suggest. Nodes make up the nodes in these layers. A neural network can have numerous hidden layers depending on the requirements. The input layer receives and passes input signals to the following layer. It collects information from the outside world. This simple type of network is interesting because the hidden units are free to construct their own representations of the input. The weights between the input and hidden units determine when each hidden unit is active, and so by modifying these weights, a hidden unit can choose what it represents. All of the calculation's back-end tasks are handled by the hidden layer. A network can have no hidden layers at all. A neural network, on the other hand, has at least one hidden layer. The output layer sends the hidden layer's calculation's ultimate result. Before giving a neural network an issue to solve, it needs some training data, just like you would with any other machine learning application. Process i. ii. iii. iv. v. vi. vii. Information is fed into the input layer which transfers it to the hidden layer The interconnections between the two layers assign weights to each input randomly A bias added to every input after weights are multiplied with them individually The weighted sum is transferred to the activation function The activation function determines which nodes it should fire for feature extraction The model applies an application function to the output layer to deliver the output Weights are adjusted, and the output is back-propagated to minimize error The model uses a cost function to reduce the error rate. You will have to change the weights with different training models. a) The model compares the output with the original result b) It repeats the process to improve accuracy The model adjusts the weights in every iteration to enhance the accuracy of the output. How neural networks learn. Neural networks must be "trained" before they can begin to function and learn on their own. They can then learn from the results they've produced and the data they've collected, but they have to start somewhere. There are a few methods that can be employed to assist neural networks in their learning process. Training:- To begin, trained neural networks are given random numbers or weights. For training, they are either supervised or unsupervised. A system that assigns a grade or corrections to the network is used in supervised training. Unsupervised training forces the network to work independently to deduce the inputs. To learn more quickly, most neural networks use supervised training. Transferred Learning:- A strategy that includes assigning a similar problem to a neural network, which may then be reused in whole or in part to speed up training and enhance performance on the task at hand. Feature Extraction:- Taking all of the data that will be given into an input, deleting any superfluous data, and grouping it into more manageable parts is what feature extraction is all about. This reduces the amount of memory and processing power required to run an issue through a neural network by just providing the network with the information it needs. Weights and Biases Some machine learning models, such as neural networks, have learnable parameters called weights and biases (abbreviated as w and b). A neural network's basic units are neurons. Each neuron in a layer of an ANN is connected to some or all of the neurons in the layer above it. When inputs are transferred across neurons, the weights and bias are applied to the inputs. The signal (or the strength of the link) between two neurons is controlled by weights. To put it another way, a weight determines how big of an impact the input has on the output. Constant biases are an additional input into the next layer that always has the value of one. The prior layer has no effect on the bias units because they have no incoming connections, but they do have outbound connections with their own weights. The bias unit ensures that even if all of the inputs are zeros, the neuron will still be activated. Applications of Neural Networks Neural Networks are regulating some key sectors including finance, healthcare, and automotive. As these artificial neurons function in a way similar to the human brain. They can be used for image recognition, character recognition and stock market predictions. Let’s understand the diverse applications of neural networks 1. Facial Recognition Facial Recognition Technologies (FRS) are effective surveillance systems. The human face is matched and compared to computer photos using Recognition Systems. For facial identification and image processing, Convolutional Neural Networks (CNN) are utilized. For the purpose of training a neural network, a large number of images are supplied into the database. The photos are then further processed for training purposes. For effective evaluations, CNN uses sampling layers. The models are fine-tuned to produce accurate recognition results. 2. Stock Market Prediction Market risks might affect investments. In the highly volatile stock market, it is practically impossible to foresee future changes. Before neural networks, the constantly changing bullish and bearish phases were unpredictable. But what altered everything? A Multilayer Perceptron MLP (class of feedforward artificial intelligence algorithm) is used to create a successful stock forecast in real time. MLP is made up of numerous layers of nodes, each of which is fully connected to the nodes above it. The MLP model is built using past stock performance, annual returns, and non profit ratios. 3. Social Media Social media has changed the mundane path of life, no matter how cliched it may sound. The study of social media user behavior employs Artificial Neural Networks. Data shared on a daily basis through virtual dialogues is tallied and analyzed for competitive purposes. The behavior of social media users is replicated by neural networks. The data can be linked to people's spending patterns after an examination of their behavior via social media networks. To mine data from social media applications, Multilayer Perceptron ANN is used. MLP uses various training methods such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Squared Error (MSE) to forecast social media trends (MSE). MLP considers a variety of parameters, such as the user's favorite Instagram pages, bookmarked options, and so on. The MLP model is trained using these factors as inputs. Artificial neural networks can surely operate as the backbone in the ever-changing dynamics of social networking apps. These factors are considered as inputs for training the MLP model. In the ever changing dynamics of social media applications, artificial neural networks can definitely work as the best fit model for user data analysis 4. Aerospace Aerospace Engineering is an expansive term that covers developments in spacecraft and aircraft. Fault diagnosis, high performance auto piloting, securing the aircraft control systems, and modeling key dynamic simulations are some of the key areas that neural networks have taken over. Time delay Neural networks can be employed for modelling non linear time dynamic systems. Time Delay Neural Networks are used for position independent feature recognition. The algorithm thus built based on time delay neural networks can recognize patterns. (Recognizing patterns are automatically built by neural networks by copying the original data from feature units). 5. Defence Defence is the backbone of every country. Every country’s state in the international domain is assessed by its military operations. Neural Networks also shape the defence operations of technologically advanced countries. The United States of America, Britain, and Japan are some countries that use artificial neural networks for developing an active defence strategy. Neural networks are used in logistics, armed attack analysis, and for object location. They are also used in air patrols, maritime patrol, and for controlling automated drones. The defence sector is getting the much needed kick of artificial intelligence to scale up its technologies. Convolutional Neural Networks(CNN), are employed for determining the presence of underwater mines. Underwater mines are the underpass that serve as an illegal commute route between two countries. Unmanned Airborne Vehicle (UAV), and Unmanned Undersea Vehicle (UUV) these autonomous sea vehicles use convolutional neural networks for the image processing. 6. Healthcare The age old saying goes like “Health is Wealth”. Modern day individuals are leveraging the advantages of technology in the healthcare sector. Convolutional Neural Networks are actively employed in the healthcare industry for X ray detection, CT Scan and ultrasound. As CNN is used in image processing, the medical imaging data retrieved from aforementioned tests is analyzed and assessed based on neural network models. Recurrent Neural Network (RNN) is also being employed for the development of voice recognition systems. Voice recognition systems are used these days to keep track of the patient’s data. Researchers are also employing Generative Neural Networks for drug discovery. Matching different categories of drugs is a hefty task, but generative neural networks have broken down the hefty task of drug discovery. They can be used for combining different elements which forms the basis of drug discovery. 7. Signature Verification and Handwriting Analysis Signature Verification , as the self explanatory term goes, is used for verifying an individual’s signature. Banks, and other financial institutions use signature verification to cross check the identity of an individual. Usually a signature verification software is used to examine the signatures. As cases of forgery are pretty common in financial institutions, signature verification is an important factor that seeks to closely examine the authenticity of signed documents. Artificial Neural Networks are used for verifying the signatures. ANN are trained to recognize the difference between real and forged signatures. ANNs can be used for the verification of both offline and online signatures. For training an ANN model, varied datasets are fed in the database. The data thus fed help the ANN model to differentiate. ANN model employs image processing for extraction of features. Handwriting analysis plays an integral role in forensics. The analysis is further used to evaluate the variations in two handwritten documents. The process of spilling words on a blank sheet is also used for behavioural analysis. Convolutional Neural Networks (CNN) are used for handwriting analysis and handwriting verification. 8. Weather Forecasting The forecasts done by the meteorological department were never accurate before artificial intelligence came into force. Weather Forecasting is primarily undertaken to anticipate the upcoming weather conditions beforehand. In the modern era, weather forecasts are even used to predict the possibilities of natural disasters. Multilayer Perceptron (MLP), Convolutional Neural Network (CNN) and Recurrent Neural Networks (RNN) are used for weather forecasting. Traditional ANN multilayer models can also be used to predict climatic conditions 15 days in advance. A combination of different types of neural network architecture can be used to predict air temperatures. Combination models (MLP+CNN), (CNN+RNN) usually works better in the case of weather forecasting. Top Companies using Artificial Neural Network(ANN) Nvidia Corp. (NVDA) Alphabet (GOOG, GOOGL) Salesforce.com (CRM) Amazon.com (AMZN) Microsoft Corp. (MSFT) Twilio (TWLO) IBM (IBM) Facebook (FB) Netflix Types of Neural Networks There are many types of neural networks available or that might be in the development stage. The three most important types of neural networks are: Artificial Neural Networks (ANN); Convolution Neural Networks (CNN), and Recurrent Neural Networks (RNN). They can be classified depending on their: Structure, Data flow, Neurons used and their density, Layers and their depth activation filters etc. A. Perceptron Perceptron model, proposed by Minsky-Papert is one of the simplest and oldest models of Neuron. It is the smallest unit of neural network that does certain computations to detect features or business intelligence in the input data. It accepts weighted inputs, and apply the activation function to obtain the output as the final result. Perceptron is also known as TLU(threshold logic unit) Perceptron is a supervised learning algorithm that classifies the data into two categories, thus it is a binary classifier. A perceptron separates the input space into two categories by a hyperplane represented by the following equation Advantages of Perceptron Perceptrons can implement Logic Gates like AND, OR, or NAND Disadvantages of Perceptron Perceptrons can only learn linearly separable problems such as boolean AND problem. For non-linear problems such as boolean XOR problem, it does not work. B. Feed Forward Neural Networks Applications on Feed Forward Neural Networks: Simple classification (where traditional Machine-learning based classification algorithms have limitations) Face recognition [Simple straight forward image processing] Computer vision [Where target classes are difficult to classify] Speech Recognition The simplest form of neural networks where input data travels in one direction only, passing through artificial neural nodes and exiting through output nodes. Where hidden layers may or may not be present, input and output layers are present there. Based on this, they can be further classified as a single-layered or multi-layered feed-forward neural network. Number of layers depends on the complexity of the function. It has uni-directional forward propagation but no backward propagation. Weights are static here. An activation function is fed by inputs which are multiplied by weights. To do so, classifying activation function or step activation function is used. For example: The neuron is activated if it is above threshold (usually 0) and the neuron produces 1 as an output. The neuron is not activated if it is below threshold (usually 0) which is considered as -1. They are fairly simple to maintain and are equipped with to deal with data which contains a lot of noise. Advantages of Feed Forward Neural Networks Less complex, easy to design & maintain Fast and speedy [One-way propagation] Highly responsive to noisy data Disadvantages of Feed Forward Neural Networks: Cannot be used for deep learning [due to absence of dense layers and back propagation] C. Multilayer Perceptron Applications on Multi-Layer Perceptron Speech Recognition Machine Translation Complex Classification An entry point towards complex neural nets where input data travels through various layers of artificial neurons. Every single node is connected to all neurons in the next layer which makes it a fully connected neural network. Input and output layers are present having multiple hidden Layers i.e. at least three or more layers in total. It has a bi-directional propagation i.e. forward propagation and backward propagation. Inputs are multiplied with weights and fed to the activation function and in backpropagation, they are modified to reduce the loss. In simple words, weights are machine learnt values from Neural Networks. They self-adjust depending on the difference between predicted outputs vs training inputs. Nonlinear activation functions are used followed by softmax as an output layer activation function. Advantages on Multi-Layer Perceptron Used for deep learning [due to the presence of dense fully connected layers and back propagation] Disadvantages on Multi-Layer Perceptron: Comparatively complex to design and maintain Comparatively slow (depends on number of hidden layers) D. Convolutional Neural Network (CNN) Convolution neural network contains a three-dimensional arrangement of neurons, instead of the standard two-dimensional array. The first layer is called a convolutional layer. Each neuron in the convolutional layer only processes the information from a small part of the visual field. Input features are taken in batch-wise like a filter. The network understands the images in parts and can compute these operations multiple times to complete the full image processing. Processing involves conversion of the image from RGB or HSI scale to grey-scale. Furthering the changes in the pixel value will help to detect the edges and images can be classified into different categories. Propagation is uni-directional where CNN contains one or more convolutional layers followed by pooling and bidirectional where the output of convolution layer goes to a fully connected neural network for classifying the images as shown in the above diagram. Filters are used to extract certain parts of the image. In MLP the inputs are multiplied with weights and fed to the activation function. Convolution uses RELU and MLP uses nonlinear activation function followed by softmax. Convolution neural networks show very effective results in image and video recognition, semantic parsing and paraphrase detection. Applications on Convolution Neural Network Image processing Computer Vision Speech Recognition Machine translation Advantages of Convolution Neural Network: Used for deep learning with few parameters Less parameters to learn as compared to fully connected layer Disadvantages of Convolution Neural Network: Comparatively complex to design and maintain Comparatively slow [depends on the number of hidden layers] E. Radial Basis Function Neural Networks Radial Basis Function Network consists of an input vector followed by a layer of RBF neurons and an output layer with one node per category. Classification is performed by measuring the input’s similarity to data points from the training set where each neuron stores a prototype. This will be one of the examples from the training set. When a new input vector [the n-dimensional vector that you are trying to classify] needs to be classified, each neuron calculates the Euclidean distance between the input and its prototype. For example, if we have two classes i.e. class A and Class B, then the new input to be classified is more close to class A prototypes than the class B prototypes. Hence, it could be tagged or classified as class A. Each RBF neuron compares the input vector to its prototype and outputs a value ranging which is a measure of similarity from 0 to 1. As the input equals to the prototype, the output of that RBF neuron will be 1 and with the distance grows between the input and prototype the response falls off exponentially towards 0. The curve generated out of neuron’s response tends towards a typical bell curve. The output layer consists of a set of neurons [one per category]. Application: Power Restoration a. Powercut P1 needs to be restored first b. Powercut P3 needs to be restored next, as it impacts more houses c. Powercut P2 should be fixed last as it impacts only one house F. Recurrent Neural Networks Designed to save the output of a layer, Recurrent Neural Network is fed back to the input to help in predicting the outcome of the layer. The first layer is typically a feed forward neural network followed by recurrent neural network layer where some information it had in the previous timestep is remembered by a memory function. Forward propagation is implemented in this case. It stores information required for it’s future use. If the prediction is wrong, the learning rate is employed to make small changes. Hence, making it gradually increase towards making the right prediction during the backpropagation. Applications of Recurrent Neural Networks Text processing like auto suggest, grammar checks, etc. Text to speech processing Image tagger Sentiment Analysis Translation Advantages of Recurrent Neural Networks Model sequential data where each sample can be assumed to be dependent on historical ones is one of the advantage. Used with convolution layers to extend the pixel effectiveness. Disadvantages of Recurrent Neural Networks Gradient vanishing and exploding problems Training recurrent neural nets could be a difficult task Difficult to process long sequential data using ReLU as an activation function. Improvement over RNN: LSTM (Long Short-Term Memory) Networks LSTM networks are a type of RNN that uses special units in addition to standard units. LSTM units include a ‘memory cell’ that can maintain information in memory for long periods of time. A set of gates is used to control when information enters the memory when it’s output, and when it’s forgotten. There are three types of gates viz, Input gate, output gate and forget gate. Input gate decides how many information from the last sample will be kept in memory; the output gate regulates the amount of data passed to the next layer, and forget gates control the tearing rate of memory stored. This architecture lets them learn longer-term dependencies This is one of the implementations of LSTM cells, many other architectures exist. G. Sequence to sequence models A sequence to sequence model consists of two Recurrent Neural Networks. Here, there exists an encoder that processes the input and a decoder that processes the output. The encoder and decoder work simultaneously – either using the same parameter or different ones. This model, on contrary to the actual RNN, is particularly applicable in those cases where the length of the input data is equal to the length of the output data. While they possess similar benefits and limitations of the RNN, these models are usually applied mainly in chatbots, machine translations, and question answering systems. H. Modular Neural Network A modular neural network has a number of different networks that function independently and perform sub-tasks. The different networks do not really interact with or signal each other during the computation process. They work independently towards achieving the output. As a result, a large and complex computational process are done significantly faster by breaking it down into independent components. The computation speed increases because the networks are not interacting with or even connected to each other. Applications of Modular Neural Network Stock market prediction systems Adaptive MNN for character recognitions Compression of high level input data Advantages of Modular Neural Network Efficient Independent training Robustness Disadvantages of Modular Neural Network Moving target Problems Advantages of Neural Networks 1) Store information on the entire network Just like it happens in traditional programming where information is stored on the network and not on a database. If a few pieces of information disappear from one place, it does not stop the whole network from functioning. 2) The ability to work with insufficient knowledge: After the training of ANN, the output produced by the data can be incomplete or insufficient. The importance of that missing information determines the lack of performance. 3) Good falt tolerance: The output generation is not affected by the corruption of one or more than one cell of artificial neural network. This makes the networks better at tolerating faults. 4) Distributed memory: For an artificial neural network to become able to learn, it is necessary to outline the examples and to teach it according to the output that is desired by showing those examples to the network. The progress of the network is directly proportional to the instances that are selected. 5) Gradual Corruption: Indeed a network experiences relative degradation and slows over time. But it does not immediately corrode the network. 6) Ability to train machine: ANN learn from events and make decisions through commenting on similar events. 7) The ability of parallel processing: These networks have numerical strength which makes them capable of performing more than one function at a time. Disadvantages of Neural Networks I. Hardware Dependence: Artificial Neural Networks require processors with parallel processing power, by their structure. For this reason, the realization of the equipment is dependent. II. Unexplained functioning of the network: This the most important problem of ANN. When ANN gives a probing solution, it does not give a clue as to why and how. This reduces trust in the network. III. Assurance of proper network structure: There is no specific rule for determining the structure of artificial neural networks. IV. The appropriate network structure is achieved through experience and trial and error. The difficulty of showing the problem to the network: ANNs can work with numerical information. Problems have to be translated into numerical values before being introduced to ANN. The display mechanism to be determined will directly influence the performance of the network. This is dependent on the user's ability. V. The duration of the network is unknown: The network is reduced to a certain value of the error on the sample means that the training has been completed. The value does not give us optimum results. Neural networks and deep learning Neural networks are a part of deep learning, which comes under the comprehensive term, artificial intelligence.Deep learning and neural networks are two terms that come to mind while thi nking about neural networks. Deep learning is essentially a very big neural network, dubbed a deep neural network. Deep neural networks feature several hidden layers, which are substantially larger than standard neural networks, allowing them to store and process more data. Deep learning and deep neural networks are a type of machine learning that uses artificial neural networks instead of only algorithms. Deep learning and deep neural networks are employed in a variety of applications nowadays; one example is chatbots that use deep resources to answer questions. Language recognition, selfdriving cars, text production, and other applications are only a few examples. Deep neural networks are essential for swiftly and successfully solving more difficult algorithms. Why neural networks are used in 21st century? Deep learning gives organizations with limitless chances to grow and improve their operations. Great changes in daily living activities can be bought through clever automation and deep learning. Although there are still arguments around AI and data ethics, businesses are increasingly relying on advanced technology as a significant resource for survival due to the numerous benefits of neural networks. Neural networks are paving the way for businesses to expand faster and function better in the face of increasing competition. Conclusion The field of neural networks is extensive. Many data scientists are only interested in neural network techniques. We barely touched on the fundamental principles in this paper. The techniques used in neural networks are far more advanced. Artificial neural networks may be thought of as simplified models of the networks of neurons that occur naturally in the animal brain. From the biological viewpoint the essential requirement for a neural network is that it should attempt to capture what we believe are the essential information processing features of the corresponding "real" network. For an engineer, this correspondence is not so important and the network offers an alternative form of parallel computing that might be more appropriate for solving the task in hand. Artificial intelligence (AI) is a very powerful and intriguing field. It will only grow in importance and scope in the future, and it will undoubtedly continue to have a huge impact on modern society. Artificial neural networks (ANNs) and the more advanced deep learning technique are among the most capable AI tools for tackling extremely complicated issues, and they will continue to be developed and used in the future. While a Terminator-style scenario is improbable in the near future, the evolution of artificial intelligence techniques and applications will be fascinating to witness! References Wikipedia Medium Section Folio3AI Blog Great Learning AI Wiki Asquero