Machine Learning Training in Coimbatore

Introduction to ML, AI

• Why we require AI and ML?
• Problem with traditional Software systems
• Opportunities with AI,ML
• What you need to excel – only a logical mind!!
• Tools and Software to efficiently build ML models
• Why R and Python(with Tensorflow) is very popular ?

Your first ML model

• Understand what actually is a ML model
• How to handle data
• Preprocessing data
• Types of ML models – Supervised and Unsupervised
• A peek into Reinforcement Learning
• How to break your data into Training and Test
• Cross validation techniques

Linear Regression

• Understand Linear Regression
• Gradient Descent
• Do actual hands on and understand the calculations behind Gradient Descent
• Brush up on Differentiation to understand the maths behind the hands-on
• Code both in R and Python
• Learn how to improve your model

Overfitting

• Overfitting is one of the most difficult aspect to learn while building a ML model
• Use the above Linear Regression model to understand Overfitting
• Learn with Hands on – how to avoid Overfitting
• Bias Variance Tradeoff
• Regularization – Ridge, LASSO
• ANOVA, F tests
• Logistic Regression
• Understand CLassification with Logistic Regression
• Maximum Likelihood Estimation
• Build an end-end model with Logistic Regression using scikit Learn
• Hands on – how actually you will build a model in the Industry
• How to code for Interviews, Data Science Competitions

Decision Trees

• What is a Probability based model and why Decision Tree is such a model? Understand the concepts of Entropy, Gini Impurity, Information Gain Do a detailed hands on project to predict the possible Loan Defaulters for a large multi-national bank
• Apply the concepts of Overfitting
• How to improve the Decision Tree model without Overfitting
• Bagging, Boosting
• Random Forest
• AdaBoost, Gradient Boost

k-NN

• Understand a Distance based model with kNN
• how to choose the value of k
• Project work on Predicting Breast Cancer

Support Vector Machines(SVM)

• The power of SVM and what it can do which other models cannot
• Why SVM is so popular in the industry
• Learn all about Kernel Functions
• Different Kernel functions
• Build an OCR(Optical Character Reader) with the help of SVM and Kernel functions
• Neural Networks
• Why Neural Networks can actually solve any Complex pattern?
• How to build the Neural Network Architecture
• How Neural Network mimics the cognitive capabilities of humans
• Build your own AND,OR,NOT,XOR,XNOR Logic Gates with Neural Network
• Understand Forward & Backward Propagation
• Plot a Neural Network with code and map your understanding between theory and practical
• Change the architecture with code and see how the Neural Network behaves
• Different Activation Functions
• Vanishing Gradient problem
• Loss functions

Deep Neural Networks

• Optimization methods
• Gradient Descent with Momentum, RMSProp, ADAM
• Learning Rate Decay
• Xavier Initialization
• Introduction to Keras and Tensorflow(TF)
• Deep Learning in Keras with TensorFlow as the backend
• Project Work

Unsupervised Learning

• Basic concepts of Clustering
• k-means Clustering
• Hierarchical clustering
• Build a hands on project to do Social Media analysis with Clustering

PCA

• Principal Component Analysis(PCA)
• Learn the maths behind PCA
• Learn how to code and plot a PCA
• Recommendation Engine
• Understand how Netflix or any other Tech giant uses Recommendation Engine
• Content and Collaborative Filtering
• Pros and Cons of different approaches of Recommendation Engine
• Market Basket Analysis
• Apriori Rules
• Build your own Recommender System

Computer Vision

• Image Detection, Image Classification, Localization
• Introduction to Convolutional Neural Networks(CNN)
• Build a Handwritten Digit recognizer with CNN
• Strides, Padding concepts
• Convolutional, Padding and Fully Connected layers
• Sliding Window
• Edge Detection

Advanced Computer Vision

• YOLO Algorithm – You Only Look Once
• Introduction to classical networks like LeNet5
• IoU
• Build an Image Classifier with CNN
• Data Augmentation Techniques
• Natural Language Processing(NLP)
• Introduction to Natural Language Processing(NLP)
• Text Preprocessing
• Lemmatization, Stemming
• Syntactical Parsing, Entity Parsing
• CTopic Modelling with Latent Dirichlet Allocation(LDA)
• Collapsed Gibbs Sampling
• Word Embedding with Word2Vec – CBOW and SkipGram models
• Restricted Boltzman Machines
• Recurrent Neural Network(RNN) and Long Short Term memory(LSTM)
• Build a chatbot with the above concepts of NLP and Neural Networks

Introduction to AI

• History of AI
• State of the Art AI
• Types of Agents
• Types of Environments
• Asymptotic Notations

Search

• Uninformed Search-Breadth first search
• Uniform Cost Search
• Depth First Search
• Depth-Limited Search
• Iterative Deepening Depth-First Search
• Bidirectional Search
• Informed Search-Greedy Best-First Search
• A* Search
• Beyond Local Search-Hill Climbing
• Simulated Annealing
• Beam Search
• Genetic Algorithms
• Online Search
• Informed Search

Adversarial Search

• Min-max Tree
• Alpha-Beta Pruning
• Move Ordering
• Stochastic games

Constraint Programming

• Constraint Satisfaction Problems
• Map coloring
• Sudoku
• Job scheduling Constraint Propagation
• Backtracking

Reinforcement learning

• Passive Reinforcement learning
• Active Reinforcement learning
• Policy Search