{"id":25526724,"date":"2022-06-04T21:29:12","date_gmt":"2022-06-04T15:59:12","guid":{"rendered":"https:\/\/entri.app\/blog\/?p=25526724"},"modified":"2022-06-04T21:29:12","modified_gmt":"2022-06-04T15:59:12","slug":"introduction-to-artificial-neutral-networks","status":"publish","type":"post","link":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/","title":{"rendered":"Introduction to Artificial Neutral Networks"},"content":{"rendered":"<div id=\"ez-toc-container\" class=\"ez-toc-v2_0_79_2 counter-hierarchy ez-toc-counter ez-toc-custom ez-toc-container-direction\">\n<p class=\"ez-toc-title\" style=\"cursor:inherit\">Table of Contents<\/p>\n<label for=\"ez-toc-cssicon-toggle-item-69e17579f3a87\" class=\"ez-toc-cssicon-toggle-label\"><span class=\"\"><span class=\"eztoc-hide\" style=\"display:none;\">Toggle<\/span><span class=\"ez-toc-icon-toggle-span\"><svg style=\"fill: #999;color:#999\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" class=\"list-377408\" width=\"20px\" height=\"20px\" viewBox=\"0 0 24 24\" fill=\"none\"><path d=\"M6 6H4v2h2V6zm14 0H8v2h12V6zM4 11h2v2H4v-2zm16 0H8v2h12v-2zM4 16h2v2H4v-2zm16 0H8v2h12v-2z\" fill=\"currentColor\"><\/path><\/svg><svg style=\"fill: #999;color:#999\" class=\"arrow-unsorted-368013\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" width=\"10px\" height=\"10px\" viewBox=\"0 0 24 24\" version=\"1.2\" baseProfile=\"tiny\"><path d=\"M18.2 9.3l-6.2-6.3-6.2 6.3c-.2.2-.3.4-.3.7s.1.5.3.7c.2.2.4.3.7.3h11c.3 0 .5-.1.7-.3.2-.2.3-.5.3-.7s-.1-.5-.3-.7zM5.8 14.7l6.2 6.3 6.2-6.3c.2-.2.3-.5.3-.7s-.1-.5-.3-.7c-.2-.2-.4-.3-.7-.3h-11c-.3 0-.5.1-.7.3-.2.2-.3.5-.3.7s.1.5.3.7z\"\/><\/svg><\/span><\/span><\/label><input type=\"checkbox\"  id=\"ez-toc-cssicon-toggle-item-69e17579f3a87\"  aria-label=\"Toggle\" \/><nav><ul class='ez-toc-list ez-toc-list-level-1 ' ><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-1\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Biological_neurons_vs_Artificial_neurons\" >Biological neurons vs Artificial neurons<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-2\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Structure_of_Artificial_neurons_and_their_functions\" >Structure of Artificial neurons and their functions<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-3\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Why_Neural_Networks\" >Why Neural Networks?<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-4\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#How_ANN_works\" >How ANN works<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-5\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Forward_Propagation\" >Forward Propagation<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-6\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Activation_functions\" >Activation functions<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-7\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Logistic_or_Sigmoid_function\" >Logistic or Sigmoid function<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-8\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Tanh_function\" >Tanh function<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-9\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#ReLU_function\" >ReLU function<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-10\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Leaky_ReLU_function\" >Leaky ReLU function<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-11\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Backpropagation\" >Backpropagation<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-12\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Terminologies\" >Terminologies:<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-13\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Coding_ANN_in_Tensorflow\" >Coding ANN in Tensorflow<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-14\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Neural_Architecture\" >Neural Architecture<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-15\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Understanding_Neural_Network\" >Understanding Neural Network<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-16\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Architecture_of_Neural_Network\" >Architecture of Neural Network<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-17\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Advantages_of_Neural_Network\" >Advantages of Neural Network<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-18\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Required_Neural_Network_Skills\" >Required Neural Network Skills<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-19\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Why_should_we_use_Neural_Networks\" >Why should we use Neural Networks?<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-20\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Neural_Networks_Scope\" >Neural Networks Scope<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-21\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#How_this_Technology_will_help_you_in_Career_Growth\" >How this Technology will help you in Career Growth<\/a><\/li><li class='ez-toc-page-1 ez-toc-heading-level-2'><a class=\"ez-toc-link ez-toc-heading-22\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#Conclusion_%E2%80%93_What_is_Neural_Networks\" >Conclusion \u2013 What is Neural Networks?<\/a><\/li><\/ul><\/nav><\/div>\n<div class=\"iw ix iy iz ja\">\n<p id=\"dc42\" class=\"pw-post-body-paragraph ma mb jd mc b md me ke mf mg mh kh mi mj mk ml mm mn mo mp mq mr ms mt mu mv iw gc\" data-selectable-paragraph=\"\">Artificial Neural Network (ANN) is a deep learning algorithm that emerged and evolved from the idea of\u00a0Biological Neural Networks of human brains. An attempt to simulate the workings of the human brain culminated in the emergence of ANN. ANN works very similar to the biological neural networks but doesn\u2019t exactly resemble its workings.<\/p>\n<p data-selectable-paragraph=\"\"><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\"><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-25520910 size-full\" src=\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square.png\" alt=\"Python and Machine Learning Square\" width=\"345\" height=\"345\" srcset=\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square.png 345w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-300x300.png 300w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-150x150.png 150w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-24x24.png 24w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-48x48.png 48w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-96x96.png 96w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-75x75.png 75w\" sizes=\"auto, (max-width: 345px) 100vw, 345px\" \/><\/a><\/p>\n<p id=\"2fa9\" class=\"pw-post-body-paragraph ma mb jd mc b md vp ke mf mg vq kh mi mj vr ml mm mn vs mp mq mr vt mt mu mv iw gc\" data-selectable-paragraph=\"\">ANN algorithm would accept only numeric and structured data as input. To accept unstructured and non-numeric data formats such as Image, Text, and Speech,\u00a0Convolutional Neural Networks (CNN),\u00a0and\u00a0Recursive Neural Networks (RNN)\u00a0are used respectively. In this post, we concentrate only on Artificial Neural Networks.<\/p>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"fe3c\" class=\"li lj jd bn lk ll lm ln lo lp lq lr ls kj lt kk lu km lv kn lw kp lx kq ly lz gc\"><span class=\"ez-toc-section\" id=\"Biological_neurons_vs_Artificial_neurons\"><\/span><strong>Biological neurons vs Artificial neurons<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3 id=\"e0ff\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\"><strong>Structure of Biological neurons and their functions<\/strong><\/h3>\n<ul class=\"\">\n<li id=\"ea7e\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\"><strong class=\"mc je\">Dendrites<\/strong>\u00a0receive incoming signals.<\/li>\n<li id=\"a612\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\"><strong class=\"mc je\">Soma<\/strong>\u00a0(cell body) is responsible for processing the input and carries biochemical information.<\/li>\n<li id=\"3e70\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\"><strong class=\"mc je\">Axon<\/strong>\u00a0is tubular in structure responsible for the transmission of signals.<\/li>\n<li id=\"fb8d\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\"><strong class=\"mc je\">Synapse<\/strong>\u00a0is present at the end of the axon and is responsible for connecting other neurons.<\/li>\n<\/ul>\n<h2 id=\"2545\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Structure_of_Artificial_neurons_and_their_functions\"><\/span><strong>Structure of Artificial neurons and their functions<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"f9e5\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">A neural network with a single layer is called a\u00a0perceptron. A multi-layer perceptron is called\u00a0Artificial Neural Networks.<\/li>\n<li id=\"70d7\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">A Neural network can possess any number of layers. Each layer can have one or more neurons or units. Each of the neurons is interconnected with each and every other neuron. Each layer could have different\u00a0activation functions\u00a0as well.<\/li>\n<li id=\"e20c\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">ANN consists of two phases\u00a0Forward propagation and Backpropagation.\u00a0The forward propagation involves multiplying weights, adding bias, and applying activation function to the inputs and propagating it forward.<\/li>\n<li id=\"37cb\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The backpropagation step is the most important step which usually involves finding optimal parameters for the model by propagating in the backward direction of the Neural network layers. The backpropagation requires\u00a0optimization function\u00a0to find the optimal weights for the model.<\/li>\n<li id=\"e03d\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">ANN can be applied to both\u00a0Regression and Classification tasks\u00a0by changing the activation functions of the output layers accordingly. (Sigmoid activation function for binary classification, Softmax activation function for multi-class classification and Linear activation function for Regression).<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Grab the opportunity to learn Python with Entri! Click Here<\/a><\/strong><\/h4>\n<h2 id=\"b5be\" class=\"li lj jd bn lk ll lm ln lo lp lq lr ls kj lt kk lu km lv kn lw kp lx kq ly lz gc\"><span class=\"ez-toc-section\" id=\"Why_Neural_Networks\"><\/span><strong>Why Neural Networks?<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"b301\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Traditional Machine Learning algorithms tend to perform at the same level when the data size increases but ANN outperforms traditional Machine Learning algorithms when the data size is huge as shown in the graph below.<\/li>\n<li id=\"9c36\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Feature Learning. The ANN tries to learn hierarchically in an incremental manner layer by layer. Due to this reason, it is not necessary to perform feature engineering explicitly.<\/li>\n<li id=\"2095\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Neural Networks can handle\u00a0unstructured data\u00a0like images, text, and speech. When the data contains unstructured data the neural network algorithms such as CNN (Convolutional Neural Networks) and RNN (Recurrent Neural Networks) are used.<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"bc2e\" class=\"li lj jd bn lk ll lm ln lo lp lq lr ls kj lt kk lu km lv kn lw kp lx kq ly lz gc\"><span class=\"ez-toc-section\" id=\"How_ANN_works\"><\/span><strong>How ANN works<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p id=\"70dc\" class=\"pw-post-body-paragraph ma mb jd mc b md me ke mf mg mh kh mi mj mk ml mm mn mo mp mq mr ms mt mu mv iw gc\" data-selectable-paragraph=\"\">The working of ANN can be broken down into two phases,<\/p>\n<ul class=\"\">\n<li id=\"cb82\" class=\"wb wc jd mc b md vp mg vq mj wq mn wr mr ws mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Forward Propagation<\/li>\n<li id=\"feb4\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Back Propagation<\/li>\n<\/ul>\n<h2 id=\"350d\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Forward_Propagation\"><\/span><strong>Forward Propagation<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"f8ef\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Forward propagation involves multiplying feature values with weights, adding bias, and then applying an activation function to each neuron in the neural network.<\/li>\n<li id=\"49d2\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Multiplying feature values with weights and adding bias to each neuron is basically applying\u00a0Linear Regression. If we apply Sigmoid function to it then each neuron is basically performing a\u00a0Logistic Regression.<\/li>\n<\/ul>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"ky kz dq la cf lb\" role=\"button\">\n<div class=\"gn go wt\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/700\/1*ZyHN5j51J92nwzdVcaFrtQ.png\" alt=\"\" width=\"700\" height=\"295\" \/><\/div>\n<\/div>\n<\/figure>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"037c\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Activation_functions\"><\/span><strong>Activation functions<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"5bbe\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The purpose of an activation function is to introduce\u00a0<strong class=\"mc je\">non-linearity<\/strong>\u00a0to the data. Introducing non-linearity helps to identify the underlying patterns which are complex. It is also used to scale the value to a particular interval. For example, the sigmoid activation function scales the value between 0 and 1.<\/li>\n<\/ul>\n<h2 id=\"656b\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Logistic_or_Sigmoid_function\"><\/span><strong>Logistic or Sigmoid function<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"c9b1\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Logistic\/ Sigmoid function scales the values between 0 and 1.<\/li>\n<li id=\"6f75\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">It is used in the output layer for Binary classification.<\/li>\n<li id=\"302f\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">It may cause a\u00a0<strong class=\"mc je\">vanishing gradient<\/strong>\u00a0problem during backpropagation and slows the training time.<\/li>\n<\/ul>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"gn go wu\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/610\/1*NSdUKVuBQt5vokwJHCwKCw.png\" alt=\"\" width=\"610\" height=\"220\" \/><\/div><figcaption class=\"le bm gp gn go lf lg bn b bo bp co\" data-selectable-paragraph=\"\">Sigmoid function<\/figcaption><\/figure>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"4fa0\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Tanh_function\"><\/span><strong>Tanh function<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"307c\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Tanh is the short form for\u00a0<strong class=\"mc je\">Hyperbolic Tangent<\/strong>. Tanh function scales the values between -1 and 1.<\/li>\n<\/ul>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"gn go wv\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/568\/1*tAE9A9tDlhnAGXq_-5y2JQ.png\" alt=\"\" width=\"568\" height=\"142\" \/><\/div><figcaption class=\"le bm gp gn go lf lg bn b bo bp co\" data-selectable-paragraph=\"\">Hyperbolic Tangent function<\/figcaption><\/figure>\n<h2 id=\"09e6\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"ReLU_function\"><\/span><strong>ReLU function<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"8b25\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\"><strong class=\"mc je\">ReLU (Rectified Linear Unit)<\/strong>\u00a0outputs the same number if x&gt;0 and outputs 0 if x&lt;0.<\/li>\n<li id=\"05f2\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">It prevents the\u00a0<strong class=\"mc je\">vanishing gradient<\/strong>\u00a0problem but introduces an\u00a0<strong class=\"mc je\">exploding<\/strong>\u00a0<strong class=\"mc je\">gradient problem<\/strong>\u00a0during backpropagation. The exploding gradient problem can be prevented by capping gradients.<\/li>\n<\/ul>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"gn go wu\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/610\/1*fALttJAfgDC1xxPgHoTpIQ.png\" alt=\"\" width=\"610\" height=\"206\" \/><\/div><figcaption class=\"le bm gp gn go lf lg bn b bo bp co\" data-selectable-paragraph=\"\">ReLU function<\/figcaption><\/figure>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"2660\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Leaky_ReLU_function\"><\/span><strong>Leaky ReLU function<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"5ec4\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Leaky ReLU is very much similar to ReLU but when x&lt;0 it returns (0.01 * x) instead of 0.<\/li>\n<li id=\"cb81\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">If the data is normalized using Z-Score it may contain negative values and ReLU would fail to consider it but leaky ReLU overcomes this problem.<\/li>\n<\/ul>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"gn go ww\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/540\/1*Uvovl617zz6eJVjQTVi3hQ.png\" alt=\"\" width=\"540\" height=\"194\" \/><\/div><figcaption class=\"le bm gp gn go lf lg bn b bo bp co\" data-selectable-paragraph=\"\">Leaky ReLU function<\/figcaption><\/figure>\n<h2 id=\"7f67\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Backpropagation\"><\/span><strong>Backpropagation<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"2aa7\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Backpropagation is done to find the\u00a0optimal value for parameters\u00a0for the model by iteratively updating parameters by partially differentiating\u00a0gradients of the loss function\u00a0with respect to the\u00a0parameters.<\/li>\n<li id=\"4e82\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">An optimization function is applied to perform backpropagation. The objective of an optimization function is to find the optimal value for parameters.<\/li>\n<\/ul>\n<p id=\"806a\" class=\"pw-post-body-paragraph ma mb jd mc b md vp ke mf mg vq kh mi mj vr ml mm mn vs mp mq mr vt mt mu mv iw gc\" data-selectable-paragraph=\"\">The optimization functions available are,<\/p>\n<ul class=\"\">\n<li id=\"9ecd\" class=\"wb wc jd mc b md vp mg vq mj wq mn wr mr ws mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Gradient Descent<\/li>\n<li id=\"c050\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Adam optimizer<\/li>\n<li id=\"6c19\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Gradient Descent with momentum<\/li>\n<li id=\"fd13\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">RMS Prop (Root Mean Square Prop)<\/li>\n<\/ul>\n<p id=\"e479\" class=\"pw-post-body-paragraph ma mb jd mc b md vp ke mf mg vq kh mi mj vr ml mm mn vs mp mq mr vt mt mu mv iw gc\" data-selectable-paragraph=\"\">The\u00a0<strong class=\"mc je\">Chain rule of Calculus<\/strong>\u00a0plays an important role in backpropagation. The formula below denotes partial differentiation of Loss (L) with respect to Weights\/ parameters (w).<\/p>\n<p id=\"3437\" class=\"pw-post-body-paragraph ma mb jd mc b md vp ke mf mg vq kh mi mj vr ml mm mn vs mp mq mr vt mt mu mv iw gc\" data-selectable-paragraph=\"\">A small change in weights \u2018w\u2019 influences the change in the value \u2018z\u2019 (\u2202\ud835\udc67\/\u2202\ud835\udc64). A small change in the value \u2018z\u2019 influences the change in the activation \u2018a\u2019 (\u2202a\/\u2202z). A small change in the activation \u2018a\u2019 influences the change in the Loss function \u2018L\u2019 (\u2202L\/\u2202a).<\/p>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"gn go wx\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/432\/1*jV3Ph2feCAOEZDi9-skl1Q.png\" alt=\"\" width=\"432\" height=\"134\" \/><\/div>\n<\/figure>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"4e52\" class=\"li lj jd bn lk ll lm ln lo lp lq lr ls kj lt kk lu km lv kn lw kp lx kq ly lz gc\"><span class=\"ez-toc-section\" id=\"Terminologies\"><\/span><strong class=\"ba\">Terminologies:<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3 id=\"9c9b\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\"><strong>Metrics<\/strong><\/h3>\n<ul class=\"\">\n<li id=\"0e2b\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">A metric is used to gauge the performance of the model.<\/li>\n<li id=\"2109\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Metric functions are similar to cost functions, except that the results from evaluating a metric are not used when training the model. Note that you may use any cost function as a metric.<\/li>\n<li id=\"c67c\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">We have used Mean Squared Logarithmic Error as a metric and cost function.<\/li>\n<\/ul>\n<figure class=\"kt ku kv kw gz kx gn go paragraph-image\">\n<div class=\"ky kz dq la cf lb\" role=\"button\">\n<div class=\"gn go wz\"><img loading=\"lazy\" decoding=\"async\" class=\"cf lc ld\" role=\"presentation\" src=\"https:\/\/miro.medium.com\/max\/700\/1*bgzceS9jybK6YzuRrGTbPw.png\" alt=\"\" width=\"700\" height=\"251\" \/><\/div>\n<\/div><figcaption class=\"le bm gp gn go lf lg bn b bo bp co\" data-selectable-paragraph=\"\">Mean Squared Logarithmic Error (MSLE) and Root Mean Squared Logarithmic Error(RMSLE)<\/figcaption><\/figure>\n<h3 id=\"ee13\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\"><strong>Epoch<\/strong><\/h3>\n<ul class=\"\">\n<li id=\"3866\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">A single pass through the training data is called an epoch. The training data is fed to the model in mini-batches and when all the mini-batches of the training data are fed to the model that constitutes an epoch.<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h3 id=\"b3e4\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\"><strong>Hyperparameters<\/strong><\/h3>\n<p id=\"f319\" class=\"pw-post-body-paragraph ma mb jd mc b md me ke mf mg mh kh mi mj mk ml mm mn mo mp mq mr ms mt mu mv iw gc\" data-selectable-paragraph=\"\">Hyperparameters are the\u00a0tunable parameters\u00a0that are not produced by a model which means the users must provide a value for these parameters. The values of hyperparameters that we provide affect the training process so hyperparameter optimization comes to the rescue.<\/p>\n<p id=\"43b1\" class=\"pw-post-body-paragraph ma mb jd mc b md vp ke mf mg vq kh mi mj vr ml mm mn vs mp mq mr vt mt mu mv iw gc\" data-selectable-paragraph=\"\">The Hyperparameters used in this ANN model are,<\/p>\n<ul class=\"\">\n<li id=\"08e4\" class=\"wb wc jd mc b md vp mg vq mj wq mn wr mr ws mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Number of layers<\/li>\n<li id=\"eded\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Number of units\/ neurons in a layer<\/li>\n<li id=\"c60c\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Activation function<\/li>\n<li id=\"9134\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Initialization of weights<\/li>\n<li id=\"9303\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Loss function<\/li>\n<li id=\"1900\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Metric<\/li>\n<li id=\"ab4c\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Optimizer<\/li>\n<li id=\"32c3\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Number of epochs<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Grab the opportunity to learn Python with Entri! Click Here<\/a><\/strong><\/h4>\n<h2 id=\"0911\" class=\"li lj jd bn lk ll lm ln lo lp lq lr ls kj lt kk lu km lv kn lw kp lx kq ly lz gc\"><span class=\"ez-toc-section\" id=\"Coding_ANN_in_Tensorflow\"><\/span><strong>Coding ANN in Tensorflow<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<h3 id=\"ba04\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\"><strong>Load the preprocessed data<\/strong><\/h3>\n<p id=\"26bf\" class=\"pw-post-body-paragraph ma mb jd mc b md me ke mf mg mh kh mi mj mk ml mm mn mo mp mq mr ms mt mu mv iw gc\" data-selectable-paragraph=\"\">The data you feed to the ANN must be preprocessed thoroughly to yield reliable results. The training data has been preprocessed already. The preprocessing steps involved are,<\/p>\n<ul class=\"\">\n<li id=\"9b91\" class=\"wb wc jd mc b md vp mg vq mj wq mn wr mr ws mv wg wh wi wj gc\" data-selectable-paragraph=\"\">MICE Imputation<\/li>\n<li id=\"4de6\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Log transformation<\/li>\n<li id=\"6ffc\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Square root transformation<\/li>\n<li id=\"52f9\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Ordinal Encoding<\/li>\n<li id=\"b635\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Target Encoding<\/li>\n<li id=\"112a\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">Z-Score Normalization<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2 id=\"b969\" class=\"vu lj jd bn lk pa vv pb lo pd vw pe ls mj vx pg lu mn vy pi lw mr vz pk ly wa gc\" data-selectable-paragraph=\"\"><span class=\"ez-toc-section\" id=\"Neural_Architecture\"><\/span><strong>Neural Architecture<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul class=\"\">\n<li id=\"93a0\" class=\"wb wc jd mc b md me mg mh mj wd mn we mr wf mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The ANN model that we are going to use, consists of seven layers including one input layer, one output layer, and five hidden layers.<\/li>\n<li id=\"5841\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The first layer (input layer) consists of 128 units\/ neurons with the ReLU activation function.<\/li>\n<li id=\"7f61\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The second, third, and fourth layers consist of 256 hidden units\/ neurons with the ReLU activation function.<\/li>\n<li id=\"3976\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The fifth and sixth layer consists of 384 hidden units with ReLU activation function.<\/li>\n<li id=\"2721\" class=\"wb wc jd mc b md wk mg wl mj wm mn wn mr wo mv wg wh wi wj gc\" data-selectable-paragraph=\"\">The last layer (output layer) consists of one single neuron which outputs an array with the shape (1, N) where N is the number of features.<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Grab the opportunity to learn Python with Entri! Click Here<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Understanding_Neural_Network\"><\/span><strong>Understanding Neural Network<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Neural networks are trained and taught just like a child\u2019s developing brain is trained. They cannot be programmed directly for a particular task. Instead, they are trained in such a manner so that they can adapt according to the changing input.<\/p>\n<p>There are three methods or learning paradigms to teach a neural network.<\/p>\n<ul>\n<li>Supervised Learning<\/li>\n<li>Reinforcement Learning<\/li>\n<li>Unsupervised Learning<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h3><strong>1. Supervised Learning<\/strong><\/h3>\n<p>As the name suggests, supervised learning means in the presence of a supervisor or a teacher. It means a set of a labeled data sets is already present with the desired output, i.e. the optimum action to be performed by\u00a0the neural network, which is already present for some data sets. The machine is then given new data sets to analyze the training data sets and to produce the correct output.<\/p>\n<p>It is a closed feedback system, but the environment is not in the loop.<\/p>\n<h3><strong>2. Reinforcement Learning<\/strong><\/h3>\n<p>In this, learning of input-output mapping is done by continuous interaction with the environment to minimise the scalar index of performance. In this method, a critic converts the primary reinforcement signal, i.e. the scalar input received from the environment, into a heuristic reinforcement signal (higher quality reinforcement signal) scalar input.<\/p>\n<p>This learning aims to minimize the cost to go function, i.e. the expected cumulative cost of actions taken over a sequence of steps.<\/p>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h3><strong>3. Unsupervised Learning<\/strong><\/h3>\n<p>As the name suggests, there is no teacher or supervisor available. In this, the data is neither labeled nor classified, and no prior guidance is available to the neural network. In this, the machine has to group the provided data sets according to the similarities, differences, and patterns without any training provided beforehand.<\/p>\n<h3><strong>Working with Neural Network<\/strong><\/h3>\n<p>The neural network is a weighted graph where nodes are the neurons, and edges with weights represent the connections. It takes input from the outside world and is denoted by x(n).<\/p>\n<p>Each input is multiplied by its respective weights, and then they are added. A bias is added if the weighted sum equates to zero, where bias has input as 1 with weight b. Then this weighted sum is passed to the activation function. The activation function limits the amplitude of the output of the neuron. There are various activation functions like the Threshold function, Piecewise linear function, or Sigmoid function.<\/p>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Architecture_of_Neural_Network\"><\/span><strong>Architecture of Neural Network<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>There are basically three types of architecture of the neural network.<\/p>\n<ul>\n<li>Single Layer feedforward network<\/li>\n<li>Multi-Layer feedforward network<\/li>\n<li>Recurrent network<\/li>\n<\/ul>\n<h4>1. Single- Layer Feedforward Network<\/h4>\n<p>In this, we have an input layer of source nodes projected on an output layer of neurons. This network is a feedforward or acyclic network. It is termed a single layer because it only refers to the computation neurons of the output layer. No computation is performed on the input layer; hence it is not counted.<\/p>\n<h4>2. Multi-Layer Feedforward Network<\/h4>\n<p>In this, there are one or more hidden layers except for the input and output layers. The nodes of this layer are called hidden neurons or hidden units. The role of the hidden layer is to intervene between the output and the external input. The input layer nodes supply the input signal to the second layer\u2019s nodes, i.e. the hidden layer, and the output of the hidden layer acts as an input for the next layer, which continues for\u00a0the rest of the network.<\/p>\n<h4>3. Recurrent Networks<\/h4>\n<p>A recurrent is almost similar to a feedforward network. The major difference is that it at least has one feedback loop. There might be zero or more hidden layers, but at least one feedback loop will be there.<\/p>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Advantages_of_Neural_Network\"><\/span><strong>Advantages of Neural Network<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Given below are the advantages mentioned:<\/p>\n<ul>\n<li>Can work with incomplete information once trained.<\/li>\n<li>Have the ability of fault tolerance.<\/li>\n<li>Have a distributed memory<\/li>\n<li>Can make machine learning.<\/li>\n<li>Parallel processing.<\/li>\n<li>Stores information on an entire network.<\/li>\n<li>Can learn non-linear and complex relationships.<\/li>\n<li>Ability to generaize, i.e. can infer unseen relationships after learning from some prior relationships.<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Grab the opportunity to learn Python with Entri! Click Here<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Required_Neural_Network_Skills\"><\/span><strong>Required Neural Network Skills<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>Given below are the required neural network skills:<\/p>\n<ul>\n<li>Knowledge of applied maths and algorithms.<\/li>\n<li>Probability and statistics.<\/li>\n<li>Distributed computing.<\/li>\n<li>Fundamental programming skills.<\/li>\n<li>Data modeling and evaluation.<\/li>\n<li>Software engineering and system design.<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Why_should_we_use_Neural_Networks\"><\/span><strong>Why should we use Neural Networks?<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<ul>\n<li>It helps to model the nonlinear and complex relationships of the real world.<\/li>\n<li>They are\u00a0used in pattern recognition\u00a0because they can generalize.<\/li>\n<li>They have many applications like text summarization, signature identification, handwriting recognition, and many more.<\/li>\n<li>It can model data with high volatility.<\/li>\n<\/ul>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Grab the opportunity to learn Python with Entri! Click Here<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Neural_Networks_Scope\"><\/span><strong>Neural Networks Scope<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>It has a wide scope in the future. Researchers are constantly working on new technologies based on neural networks. Everything is converting into automation; hence they are very much efficient in dealing with changes and can adapt accordingly. Due to an increase in new technologies, there are many job openings for engineers and neural network experts. Hence in the future also neural networks will prove to be a major job provider.<\/p>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Ace your coding skills with Entri !<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"How_this_Technology_will_help_you_in_Career_Growth\"><\/span><strong>How this Technology will help you in Career Growth<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>There is huge career growth in the field of neural networks. The average salary of a neural network engineer ranges from $33,856 to $153,240 per year approximately.<\/p>\n<h4 style=\"text-align: center\"><strong><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\" target=\"_blank\" rel=\"noopener\">Grab the opportunity to learn Python with Entri! Click Here<\/a><\/strong><\/h4>\n<h2><span class=\"ez-toc-section\" id=\"Conclusion_%E2%80%93_What_is_Neural_Networks\"><\/span><strong>Conclusion \u2013 What is Neural Networks?<\/strong><span class=\"ez-toc-section-end\"><\/span><\/h2>\n<p>There is a lot to gain from neural networks. They can learn and adapt according to the changing environment. Moreover, they contribute to other areas as well as in the field of neurology and psychology. Hence there is a huge scope of neural networks in today\u2019s time as well as in the future.<\/p>\n<p><a href=\"https:\/\/entri.sng.link\/Bcofz\/yeoy\/ojyv\"><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter wp-image-25520910 size-full\" src=\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square.png\" alt=\"Python and Machine Learning Square\" width=\"345\" height=\"345\" srcset=\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square.png 345w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-300x300.png 300w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-150x150.png 150w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-24x24.png 24w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-48x48.png 48w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-96x96.png 96w, https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/04\/Python-and-Machine-Learning-Square-75x75.png 75w\" sizes=\"auto, (max-width: 345px) 100vw, 345px\" \/><\/a><\/p>\n<figure class=\"kt ku kv kw gz kx\">\n<div class=\"m l dq\">\n<div class=\"abm ro l\"><\/div>\n<\/div>\n<\/figure>\n<\/div>\n<div class=\"o dz te xa ii xb\" role=\"separator\"><\/div>\n","protected":false},"excerpt":{"rendered":"<p>Artificial Neural Network (ANN) is a deep learning algorithm that emerged and evolved from the idea of\u00a0Biological Neural Networks of human brains. An attempt to simulate the workings of the human brain culminated in the emergence of ANN. ANN works very similar to the biological neural networks but doesn\u2019t exactly resemble its workings. ANN algorithm [&hellip;]<\/p>\n","protected":false},"author":111,"featured_media":25526954,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[802,1864],"tags":[],"class_list":["post-25526724","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-articles","category-data-science-ml"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.6 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Introduction to Artificial Neutral Networks - Entri Blog<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Introduction to Artificial Neutral Networks - Entri Blog\" \/>\n<meta property=\"og:description\" content=\"Artificial Neural Network (ANN) is a deep learning algorithm that emerged and evolved from the idea of\u00a0Biological Neural Networks of human brains. An attempt to simulate the workings of the human brain culminated in the emergence of ANN. ANN works very similar to the biological neural networks but doesn\u2019t exactly resemble its workings. ANN algorithm [&hellip;]\" \/>\n<meta property=\"og:url\" content=\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\" \/>\n<meta property=\"og:site_name\" content=\"Entri Blog\" \/>\n<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/entri.me\/\" \/>\n<meta property=\"article:published_time\" content=\"2022-06-04T15:59:12+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png\" \/>\n\t<meta property=\"og:image:width\" content=\"820\" \/>\n\t<meta property=\"og:image:height\" content=\"615\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"author\" content=\"Feeba Mahin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@entri_app\" \/>\n<meta name=\"twitter:site\" content=\"@entri_app\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Feeba Mahin\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"11 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\"},\"author\":{\"name\":\"Feeba Mahin\",\"@id\":\"https:\/\/entri.app\/blog\/#\/schema\/person\/f036dab84abae3dcc9390a1110d95d36\"},\"headline\":\"Introduction to Artificial Neutral Networks\",\"datePublished\":\"2022-06-04T15:59:12+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\"},\"wordCount\":2236,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/entri.app\/blog\/#organization\"},\"image\":{\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png\",\"articleSection\":[\"Articles\",\"Data Science and Machine Learning\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\",\"url\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\",\"name\":\"Introduction to Artificial Neutral Networks - Entri Blog\",\"isPartOf\":{\"@id\":\"https:\/\/entri.app\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage\"},\"image\":{\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage\"},\"thumbnailUrl\":\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png\",\"datePublished\":\"2022-06-04T15:59:12+00:00\",\"breadcrumb\":{\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage\",\"url\":\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png\",\"contentUrl\":\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png\",\"width\":820,\"height\":615,\"caption\":\"Introduction to Artificial Neutral Network\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/entri.app\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Entri Skilling\",\"item\":\"https:\/\/entri.app\/blog\/category\/entri-skilling\/\"},{\"@type\":\"ListItem\",\"position\":3,\"name\":\"Data Science and Machine Learning\",\"item\":\"https:\/\/entri.app\/blog\/category\/entri-skilling\/data-science-ml\/\"},{\"@type\":\"ListItem\",\"position\":4,\"name\":\"Introduction to Artificial Neutral Networks\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/entri.app\/blog\/#website\",\"url\":\"https:\/\/entri.app\/blog\/\",\"name\":\"Entri Blog\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\/\/entri.app\/blog\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/entri.app\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\/\/entri.app\/blog\/#organization\",\"name\":\"Entri App\",\"url\":\"https:\/\/entri.app\/blog\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/entri.app\/blog\/#\/schema\/logo\/image\/\",\"url\":\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2019\/10\/Entri-Logo-1.png\",\"contentUrl\":\"https:\/\/entri.app\/blog\/wp-content\/uploads\/2019\/10\/Entri-Logo-1.png\",\"width\":989,\"height\":446,\"caption\":\"Entri App\"},\"image\":{\"@id\":\"https:\/\/entri.app\/blog\/#\/schema\/logo\/image\/\"},\"sameAs\":[\"https:\/\/www.facebook.com\/entri.me\/\",\"https:\/\/x.com\/entri_app\"]},{\"@type\":\"Person\",\"@id\":\"https:\/\/entri.app\/blog\/#\/schema\/person\/f036dab84abae3dcc9390a1110d95d36\",\"name\":\"Feeba Mahin\",\"url\":\"https:\/\/entri.app\/blog\/author\/feeba123\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Introduction to Artificial Neutral Networks - Entri Blog","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/","og_locale":"en_US","og_type":"article","og_title":"Introduction to Artificial Neutral Networks - Entri Blog","og_description":"Artificial Neural Network (ANN) is a deep learning algorithm that emerged and evolved from the idea of\u00a0Biological Neural Networks of human brains. An attempt to simulate the workings of the human brain culminated in the emergence of ANN. ANN works very similar to the biological neural networks but doesn\u2019t exactly resemble its workings. ANN algorithm [&hellip;]","og_url":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/","og_site_name":"Entri Blog","article_publisher":"https:\/\/www.facebook.com\/entri.me\/","article_published_time":"2022-06-04T15:59:12+00:00","og_image":[{"width":820,"height":615,"url":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png","type":"image\/png"}],"author":"Feeba Mahin","twitter_card":"summary_large_image","twitter_creator":"@entri_app","twitter_site":"@entri_app","twitter_misc":{"Written by":"Feeba Mahin","Est. reading time":"11 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#article","isPartOf":{"@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/"},"author":{"name":"Feeba Mahin","@id":"https:\/\/entri.app\/blog\/#\/schema\/person\/f036dab84abae3dcc9390a1110d95d36"},"headline":"Introduction to Artificial Neutral Networks","datePublished":"2022-06-04T15:59:12+00:00","mainEntityOfPage":{"@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/"},"wordCount":2236,"commentCount":0,"publisher":{"@id":"https:\/\/entri.app\/blog\/#organization"},"image":{"@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage"},"thumbnailUrl":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png","articleSection":["Articles","Data Science and Machine Learning"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/","url":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/","name":"Introduction to Artificial Neutral Networks - Entri Blog","isPartOf":{"@id":"https:\/\/entri.app\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage"},"image":{"@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage"},"thumbnailUrl":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png","datePublished":"2022-06-04T15:59:12+00:00","breadcrumb":{"@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#primaryimage","url":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png","contentUrl":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2022\/06\/Introduction-to-Artificial-Neutral-Network.png","width":820,"height":615,"caption":"Introduction to Artificial Neutral Network"},{"@type":"BreadcrumbList","@id":"https:\/\/entri.app\/blog\/introduction-to-artificial-neutral-networks\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/entri.app\/blog\/"},{"@type":"ListItem","position":2,"name":"Entri Skilling","item":"https:\/\/entri.app\/blog\/category\/entri-skilling\/"},{"@type":"ListItem","position":3,"name":"Data Science and Machine Learning","item":"https:\/\/entri.app\/blog\/category\/entri-skilling\/data-science-ml\/"},{"@type":"ListItem","position":4,"name":"Introduction to Artificial Neutral Networks"}]},{"@type":"WebSite","@id":"https:\/\/entri.app\/blog\/#website","url":"https:\/\/entri.app\/blog\/","name":"Entri Blog","description":"","publisher":{"@id":"https:\/\/entri.app\/blog\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/entri.app\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/entri.app\/blog\/#organization","name":"Entri App","url":"https:\/\/entri.app\/blog\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/entri.app\/blog\/#\/schema\/logo\/image\/","url":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2019\/10\/Entri-Logo-1.png","contentUrl":"https:\/\/entri.app\/blog\/wp-content\/uploads\/2019\/10\/Entri-Logo-1.png","width":989,"height":446,"caption":"Entri App"},"image":{"@id":"https:\/\/entri.app\/blog\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.facebook.com\/entri.me\/","https:\/\/x.com\/entri_app"]},{"@type":"Person","@id":"https:\/\/entri.app\/blog\/#\/schema\/person\/f036dab84abae3dcc9390a1110d95d36","name":"Feeba Mahin","url":"https:\/\/entri.app\/blog\/author\/feeba123\/"}]}},"_links":{"self":[{"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/posts\/25526724","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/users\/111"}],"replies":[{"embeddable":true,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/comments?post=25526724"}],"version-history":[{"count":6,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/posts\/25526724\/revisions"}],"predecessor-version":[{"id":25526959,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/posts\/25526724\/revisions\/25526959"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/media\/25526954"}],"wp:attachment":[{"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/media?parent=25526724"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/categories?post=25526724"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/entri.app\/blog\/wp-json\/wp\/v2\/tags?post=25526724"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}