Supervised Machine Learning - Basics.

• Welcome to machine learning!
• Applications of machine learning
• What is machine learning?
• Supervised learning
• Unsupervised learning
• Jupyter Notebooks
• Linear regression model
• Cost function
• Visualizing the cost function
• Gradient descent
• Learning rate
• Gradient descent for linear regression

Supervised Machine Learning - Regression with multiple input variables.

• Multiple features
• Vectorization
• Gradient descent for multiple linear regression
• Feature scaling
• Checking gradient descent for convergence
• Choosing the learning rate
• Feature engineering
• Polynomial regression

Supervised Machine Learning - Classification

• Motivations
• Logistic regression
• Decision boundary
• Cost function for logistic regression
• Gradient Descent Implementation
• Overfitting
• Cost function with regularization
• Regularized linear regression
• Regularized logistic regression

Neural Network

• Neurons and the brain
• Demand Prediction
• Neural network layer
• More complex neural networks
• Inference: making predictions (forward propagation)
• Inference in Code
• Data in TensorFlow
• Building a neural network
• Forward propagation in a single layer
• General implementation of forward propagation
• How neural networks are implemented efficiently
• Matrix multiplication

Neural Network - Implementation

• TensorFlow implementation
• Training Details
• Activation functions
• Multiclass
• Softmax
• Neural Network with Softmax output
• Classification with multiple outputs
• Advanced Optimization
• Additional Layer Types
• Larger neural network example
• Evaluating a model
• Model selection and training/cross validation/test sets
• Diagnosing bias and variance
• Regularization and bias/variance
• Establishing a baseline level of performance
• Learning curves
• Bias/variance and neural networks
• Iterative loop of ML development
• Error analysis
• Adding data
• Transfer learning: using data from a different task
• Full cycle of a machine learning project
• Fairness, bias, and ethics
• Error metrics for skewed datasets
• Trading off precision and recall

Decision Trees

• Decision tree model
• Learning Process
• Measuring purity
• Choosing a split: Information Gain
• Putting it together
• Using one-hot encoding of categorical features
• Continuous valued features
• Regression Trees
• Using multiple decision trees
• Sampling with replacement
• Random forest algorithm
• When to use decision trees

Unsupervised Learning

• What is clustering?
• K-means algorithm
• Optimization objective
• Initializing K-means
• Choosing the number of clusters
• Finding unusual events
• Gaussian (normal) distribution
• Anomaly detection algorithm
• Developing and evaluating an anomaly detection system
• Anomaly detection vs. supervised learning
• Choosing what features to use

Neural Network and Deep Network

• What is a Neural Network?
• Computing a Neural Network's Output
• Activation Functions
• Why do you need Nonlinear Activation Functions?
• Gradient Descent for Neural Networks
• Random Initialization
• Supervised Learning with Neural Networks•8 minutes
• Why is Deep Learning taking off?
• Deep L-layer Neural Network•5 minutes
• Forward Propagation in a Deep Network
• Getting your Matrix Dimensions Right
• Why Deep Representations?
• Building Blocks of Deep Neural Networks
• Forward and Backward Propagation
• Parameters vs Hyperparameters
• What does this have to do with the brain?

Convolutional Neural Network

• Edge Detection Example
• More Edge Detection
• Padding
• Strided Convolutions
• Convolutions Over Volume
• One Layer of a Convolutional Network
• Simple Convolutional Network Example
• Pooling Layers
• CNN Example
• Why Convolutions?

Sequence Models - Recurrent Neural Network

• Why Sequence Models?
• Recurrent Neural Network Model
• Backpropagation Through Time
• Different Types of RNNs
• Language Model and Sequence Generation
• Sampling Novel Sequences
• Vanishing Gradients with RNNs
• Gated Recurrent Unit (GRU)
• Long Short Term Memory (LSTM)
• Bidirectional RNN
• Deep RNNs

Practical Aspects in Deep Network

• Train / Dev / Test sets
• Bias / Variance
• Basic Recipe for Machine Learning
• Regularization
• Why Regularization Reduces Overfitting?
• Dropout Regularization
• Understanding Dropout
• Other Regularization Methods
• Normalizing Inputs
• Vanishing / Exploding Gradients
• Weight Initialization for Deep Networks
• Numerical Approximation of Gradients
• Gradient Checking

Optimization Algorithms

• Mini-batch Gradient Descent
• Exponentially Weighted Averages
• Gradient Descent with Momentum
• RMSprop
• Adam Optimization Algorithm
• Learning Rate Decay
• The Problem of Local Optima

Hyperparameter Tuning

• Tuning Process
• Using an Appropriate Scale to pick Hyperparameters
• Normalizing Activations in a Network
• Batch Normalization
• Softmax Regression
• Deep Learning Frameworks

Tensorflow for AI

• The ‘Hello World’ of neural networks
• Working through ‘Hello World’ in TensorFlow and Python
• Writing code to load training data
• Coding a Computer Vision Neural Network
• Implementing convolutional layers
• Implementing pooling layers
• Develop a Classifier project

Executing Deep Codes

• System Configuration - Hardware Requirements
• System Configuration - Software Requirements
• Using Google Colab, H2O, etc.

Python for ML

• Installation Steps for Windows/Linux/MAC
• Introduction to Python
• Introduction to Jupyter Notebooks
• Variables and Data types, Data Structures, Operators
• Conditional Statements, Looping Constructs
• String Manipulation, Functions
• Moducles, Packages and Standard Libraries
• Handling text files in Python
• Introduction to Python Libraries for Data Science
• Numpy
• Scipy
• Pandas
• Scikit-learn
• Matplotlib
• Statsmodels
• Reading Data files in Python
• CSV files
• JSON files
• Reading Excel & Spreadsheet files