Machine Learning

Prerequisites for the course: Working knowledge of Python programming, undergraduate mathematics, and CLI experience.

Emphasis will be on hands on development, going through exercises and workshops together. Very little reading.

Cost of the Course: $5,000

1. Machine Learning Overview
   1. Introduction
   2. Uses for Machine Learning
   3. Environment configuration for Python Development with Anaconda IDE
   4. Python Library Imports
2. Preprocessing
   1. Importance of data management
   2. Difference Between Independent and Dependent Variables in Data Science
   3. How to Import Data and Segment I and D Variables
   4. Script development for cleaning data
      1. Object Inspection
      2. Replacing NaN Cells in Python with the Mean, Median and Mode
   5. Missing value management
   6. Setting up a Test and Training Set of Data
   7. Normalization and Feature Scaling
   8. Data categorization
   9. Numpy
      1. Arrays
         1. Dtype
         2. Upcasting: Implicit Casting, Astype, Structured Array
         3. Copy function
         4. Nan for placeholder element
         5. Inf for infinity
         6. transposing
3. Regression
   1. Linear regression
      1. 5 Assumptions of a Linear Regression
      2. Simple
      3. Multiple
      4. Polynomial
   2. Decision Trees
   3. Forests
4. Classification
   1. Naive bayes
      1. Generative vs Discriminative Models
      2. Spam classification system
   2. K Nearest Neighbors
      1. Visualizations
      2. XOR Problem
   3. Decision Trees
      1. Information Entropy
      2. Maximizing Information Gain
   4. Random forest
   5. Logistic regressions
   6. Perceptron Concepts
   7. Bias-Variance
   8. Adaboost
5. Bayesian Networks
   1. Overview
   2. Probability Management
   3. Exploring and Exploiting
   4. Threshold Management
   5. Building an A/B Testing system
   6. Multivariate Gaussian Likelihood
6. Clustering
   1. K Means Clustering
   2. Hierarchy based clustering
   3. Cluster visualizations
7. Neural Networks
   1. Overview
   2. Tensorflow
   3. Image recognition
   4. Boltzmann Machines
   5. Intuition Development
   6. Convolutional vs Artificial vs Recurrent
   7. LeNet
   8. Theano
   9. Facial Recognition Engine
   10. Convolutional Neural Networks (CNN)
   11. Artificial Neural Networks (ANN)
   12. Recurrent Neural Networks (RNN)
8. Vendor Machine Learning Systems
   1. Google NLP engine
   2. AWS NLU (Natural Language Understanding)
   3. AWS Predictive Models
9. Special Machine Learning Problem Case Studies
   1. Markov Decision Processes
   2. Rebuilding Google PageRank
   3. Monte Carlo
   4. Temporal Difference Learning
   5. Approximation Methods