GANs consist of three terms Generative Adversarial Network. Let's understand these three terms first. Generative : A Generative Model takes input training sample from some distribution and learns to represents that distribution. Adversarial : It basically means Conflicting or Opposing. Networks : These are basically neural networks. So,Generative Adversarial Networks are deep neural network architecture comprising of two neural networks compete with each other to make a generative model. A GAN consist of two class models : Discriminative Model :- It is the one that discriminate between two different classes of data.It tries to identify real data from fakes created by the generator Generative Model :- The Generator turns noise into an imitation of the data to try to trick the discriminator Mathematically, A Generative Model 'G' to be trained on training data 'X' sampled from some true distribution 'D' is the one which, given some standard random distrib...
What is Machine Learning?
Machine Learning is the study of computer algorithms that allow computer programs to automatically improve through experience.Machine learning (ML) is a type of artificial intelligence (AI) that allows software applications to become more accurate at predicting outcomes without being explicitly programmed to do so. Machine learning algorithms use historical data as input to predict new output values.
1. Learn Programming
Choose any programming language( Generally Python/R) and learn interactively and also learn the Fundamentals to Computer Science.
Resources
2. Linear Algebra
Lot of ML concepts are tied up with linear algebra. It also helps us to understand concepts like PCA and also you can build algorithms from scratch. Some important key concepts to learn are Matrices, Eigenvectors & Eigenvalues, Jacobian Matrix, Gradient Decent and Curse of Dimentionality and matrix decompositions.
Resource
mml-book.com
3. Statistics and Probability
Learn Descriptive and inferential, associative, and differential statistical concepts and also probability ( Conditional Probability and Bayes Theorem). For some important key concepts check out Why Probability & Statistics.
Resources
Introduction to statistical learning or elements of statistical learning
4.Calculus and Optimization
Calculus is very important concept if you want to understand how optimization works in Machine Learning and Deep Learning. It is very important if your focus in on research side. Some important key concepts to learn are Differentiability, Maxima & Minima, product & chain rule, Gradient Decent,SGD, mini DGD, Momentum, Adagrad etc.
Resources
Thomas calculus
5.Data Preprocessing
It is very important step where we analyse data to extract patters wo make business decisions. This step also helps to convert the data into right format and fix if any errors. we can also create/remove features.Some important key concepts are Data types, data exploration, Feature cleaning/imputation/engineering/selection/encoding,scaling etc
Resources
6.Understand Loss Function
Loss Function or Cost Function is a function that maps an event or values of one or more variables onto a real number intutively representing some 'cost' associated with the event. Some key concepts Maximum likelihood estimation(MLE), Cross-Entropy, Logistic Loss, 0-1 Loss, KL Divergence etc.
Resources
7.Machine Learning
To Start Machine Learning , first understand the terminologies around machine learning and its types (supervised learning, unsupervised learning, Dimensionality reduction techniques, time series etc. Some key concepts to learn are Different types of learning, parameter types, understand ML taxonomy, Different types of algorithms etc
Resources
Hands-On Machine Learning with scikit-learn and TensorFlow
8.Performance Analysis
Model Evaluation helps to find the best model that represents out data and how well the choosen model work in the future. Some key concepts to learn Confusion Matrix, Accuracy, f1 score, AUC ROC, Bias - Variance tradeoff , MSE, ErrorRate etc
Resources
9.Hyper Parameter Tuning
These are not model parameters and cannot trained directly from the data. It is a process of choosing a set of optimal parameters to control the learning process. Some key concepts to learn are cross validation, Different hyper parameter tuning techniques, Early stopping, Overfiting, Underfiting etc.
Resources
10.Learn Regularization
Regularisation is a technique used to reduce the errors by fitting the function appropriately on the given training set and avoid overfitting. Some key concepts are L1 norm, L2 norm, Early stopping, sparse regularization, Elastic net, Mean-Constrained etc
Resources
Get a Job
Best way to test your skills is to work on life problem. you can get a job/internship or take a bootcamp or work on your own venture where you can do this. Start with an internship, Bootcamp and get a full-time.
Comments
Post a Comment