3
127
Unlimited Duration
March 9, 2023
Courses
Students
Machine learning is a branch of artificial intelligence (AI) and computer science which focuses on the use of data and algorithms to imitate the way that humans learn, gradually improving its accuracy.Artificial intelligence (AI) and machine learning are often used interchangeably, but machine learning is a subset of the broader category of AI.
Put in context, artificial intelligence refers to the general ability of computers to emulate human thought and perform tasks in realworld environments, while machine learning refers to the technologies and algorithms that enable systems to identify patterns, make decisions, and improve themselves through experience and data.
Computer programmers and software developers enable computers to analyze data and solve problems â€” essentially, they create artificial intelligence systems â€” by applying tools such as:
 Machine learning
 Deep learning
 Neural networks
 Computer vision
 Natural language processing

 What is machine learning? Unlimited
 Why this Course? Unlimited
 What is the difference between artificial intelligence and machine learning? Unlimited
 Humans use the rememberformulate Unlimited
 How do humans think? Unlimited
 How do machines think? Unlimited
 Summary Unlimited
 Module 1. What is AI and machine learning Deck Unlimited

 Types of machine learning Unlimited
 What is the difference between labelled and unlabelled data? Unlimited
 What is supervised learning? Unlimited
 Regression models predict numbers Unlimited
 Classification models predict a state Unlimited
 What is unsupervised learning? Unlimited
 Clustering algorithms split a dataset into similar groups Unlimited
 Dimensionality reduction simplifies data Unlimited
 What is reinforcement learning? Unlimited
 Summary Unlimited

 Linear regression Unlimited
 The problem Unlimited
 The remember step Unlimited
 the linear regression algorithm Unlimited
 A simple trick Unlimited
 The square trick Unlimited
 Using the linear regression algorithm in our dataset Unlimited
 Summary Unlimited

 Optimizing the training process Unlimited
 The model evaluation graph Unlimited
 Regularization Unlimited
 Movie recommendations Unlimited
 Measuring how complex a model Unlimited
 The Regularization parameter Unlimited
 An intuitive way to see regularization Unlimited
 Summary Unlimited

 The perceptron algorithm Unlimited
 The problem Unlimited
 Simple sentiment analysis classifier Unlimited
 A slightly more complicated planet Unlimited
 More general cases Unlimited
 The error function Unlimited
 The perceptron trick Unlimited
 The perceptron algorithm Unlimited
 Coding the perceptron trick Unlimited
 Applications of the perceptron algorithm Unlimited
 Summary Unlimited

 Logistic regression Unlimited
 The sigmoid function Unlimited
 The error functions Unlimited
 More on the log loss error function Unlimited
 An example with a discrete perceptron & continuous perceptron Unlimited
 Coding the logistic regression algorithm Unlimited
 The softmax function Unlimited
 Summary Unlimited

 Accuracy and its friends Unlimited
 Coronavirus and spam email Unlimited
 How to fix the accuracy problem? Unlimited
 The confusion matrix Unlimited
 how many did we correctly classify? Unlimited
 CALCULATING THE F SCORE Unlimited
 A metric that tells us how good our model is Unlimited
 the ROC curve Unlimited
 Summary Unlimited

 The naive Bayes algorithm Unlimited
 A story with Bayes Theorem Unlimited
 Spam detection model Unlimited
 The probability that any email is spam Unlimited
 What the math just happened? Unlimited
 What about two words? Unlimited
 Building a spam detection model Unlimited
 Data preprocessing Unlimited
 Finding the priors Unlimited
 Finding the posteriors with Bayes theorem Unlimited
 Summary Unlimited

 Decision trees Unlimited
 The problem Unlimited
 First step to build the model Unlimited
 Next and final step Unlimited
 Building the tree Unlimited
 Accuracy Unlimited
 Gini impurity Unlimited
 Building our decision tree using Gini index Unlimited
 Features with more categories, such as Dog/Cat/Bird Unlimited
 Continuous features, such as a number Unlimited
 Coding a decision tree with sklearn Unlimited
 A slightly larger example Unlimited
 Summary Unlimited

 Neural Networks Unlimited
 The problem Unlimited
 Perceptrons and how to combine them Unlimited
 A trick to improve our training Unlimited
 The architecture of a neural network Unlimited
 Training neural networks Unlimited
 How to code a neural network in Keras Unlimited
 Other more complicated architectures and some scifi applications Unlimited
 Summary Unlimited

 Combining models to maximize results Ensemble learning Unlimited
 Why an ensemble of learners? Unlimited
 Building random forests by joining several trees Unlimited
 Coding a random forest in sklearn Unlimited
 A big picture of AdaBoost Unlimited
 A detailed (mathematical) picture of AdaBoost Unlimited
 Coding AdaBoost in Sklearn Unlimited
 Summary Unlimited

 Support vector machines and the kernel method Unlimited
 A new error function Unlimited
 Classification error function Unlimited
 Distance error function Unlimited
 The C parameter Unlimited
 Coding a simple SVM Unlimited
 The kernel method Unlimited
 The radial basis function (rbf) kernel Unlimited
 Radial basis functions Unlimited
 The gamma parameter Unlimited
 Summary Unlimited

 Putting it in practice Unlimited
 The features of our dataset Unlimited
 Using pandas to load the dataset Unlimited
 Cleaning up our dataset Unlimited
 Dropping columns with missing data Unlimited
 Turning categorical data into numerical data Unlimited
 Binning Unlimited
 Saving for future use Unlimited
 Training our models Unlimited
 TESTING EACH MODELâ€™S ACCURACY Unlimited
 Grid search Unlimited
 Using Kfold crossvalidation Unlimited
 Summary Unlimited