INTRODUCTION TO DEEPLEARNING BASICS(TRANSITION FROM MACHINE LEARNING TO DEEP LEARNING)
TABLE OF CONTENTS¡Deep Learning:What and Why¡Key differences between ML and DL¡Basic Terminologies¡Perceptron¡Various Deep Learning Networks¡Transfer Learning¡Applications of Deep Learning
DEEP LEARNING¡Subfield of Machine Learning¡Concerned with algorithms inspired by the structure and function of the brain calledartificial neural networks¡Attempts to draw similar conclusions as humans would by continuously analyzingdata with a given logical structure
DL ⊂ ML ⊂ ࠵?࠵?
DEEP LEARNING OVER MACHINE LEARNING¡Conventional ML methods: succumb to environmental changes¡Deep learning:Adapts to changes by constant feedback and improve the model¡Deep Learning can be used to solve any complex problem
BASIC TERMINOLOGIES (CONTD.)¡Weights:When input enters the neuron, it is multiplied by a weightInitialization of weights is done randomlyUpdated during the training processMore weight: More important feature¡Bias:Constant added to the NN to change the range ofmultiplied input¡Activation Function:Used to “fire” a neuronLike a gate that checks that an incoming value is greaterthan a critical numberNote:A neural network without an activation function is essentially just a linear regression model.