– Statistical Learning
– Python For AI and ML

– Uninformed and heuristic-based search techniques
– Adversarial search and its uses
– Planning and constraint satisfaction techniques

– Neural Network Basics
– Deep Neural Networks
– Recurrent Neural Networks (RNN)
– Deep Learning applied to images using CNN
– Tensor Flow for Neural Networks & Deep Learning

– Convolutional Neural Networks
– Keras library for deep learning in Python
– Pre-processing image Data
– Object & face recognition

– Visualizing features & kernels
– TensorBoard – Visualizing Learning, Graph Visualization
– Synthesis and style transfer
– Case Study: Visualizing a convoluted neural network

– Statistical NLP and text similarity
– Syntax and parsing techniques
– Text summarization techniques
– Semantics and Generation

– Introduction to machine learning
– Key terminologies
– Key tasks of machine learning
– Choosing a right algorithm
– Application development steps

– Classifying with distance measurement
– Improving matches from a dating site using KNN
– Handwriting recognition system

– Tree construction
– Plotting trees in Python with matplottlib annotation
– Testing and storing a classifier
– Using decision tree to predict contact lenses type.

– Classifying with Bayesian decision theory
– Conditional Probability
– Classifying with conditional Probabilities
– Documents classification with naïve bayes
– Classifying text with Python
– Classifying Naïve Bayes with Python

– Classification with logistic regression and the sigmoid function: a tractable step function
– Using optimization to find the best regression coefficient
– Estimating horse fatalities from colic database

– Separating data with a maximum margin
– Finding a maximum margin
– Efficient optimization with SMO algorithm
– Speeding up optimization with full Platt SMO
– Using kernels for more complex data
– Revisiting handwriting classification

– Classifiers using multiple samples of dataset
– Train: Improving the classifier by focussing on errors
– Creating a weak learner with descision stump
– Implementing the full Adaboost algorithm
– Test: Classifying with AdaBoost
– Adaboost on difficult dataset

– Finding best-fir lines with linear regression
– Locally weighted linear regression
– Predicting the age of abalone
– Shrinking coefficient to understand data
– The bias/variance trade off
– Forecasting the price of LEGO sets

– The k-Means clustering algorithm
– Improving clustering performance with post processing
– Bisecting K-means
– Clustering points on a map

– Association analysis
– The Apriory principle
– Finding frequent itemsets with the Apriori algorithm
– Mining association rules from frequent itemsets
– Uncovering patterns in congressional votings
– Finding similar features in poisonous mushrooms

– FP-trees: an efficient way to encode a dataset
– Build an FP tree
– Mining frequent items from an FP tree
– Finding co-occurring words in twitter field
– Mining a clickstream from a news site