Deep Learning using Python

About course
Deep Learning is one of the exciting & fast growing fields within data science. One question that practicing data scientists regularly face from business users is - "What is Deep Learning and how is it different from the traditional data science algorithms that some companies have already been using for some time".
This is a 5 week course designed to bridge the gap between theory and practice. While most courses in the market are focused on machine learning algorithms (such as gradient boosting and random forest), this course is one of its kind that takes a case study approach in covering the implementation of the latest deep learning techniques - Convolutional/ Recurrent Neural networks & auto encoders in processing text, audio & image datasets.

Week 1
Machine learning - key concepts, Model Scoping
Machine learning vs. AI, Learning Mechanisms, Data Preparation, Model Evaluation/Selection, Bias-variance trade-off
Python Basics
Overview, Basics of dataframes, Lists, Arrays, Loops, Pandas/Numpy
Week 2
Building a Neural Network in Excel
Neural Network concepts - Activation, Error functions, Back-Propagation, Optimization, Key Challenges
Introduction to Keras
Builiding feed-forward in keras and other details
Week 3
Fine-tuning Neural Networks
Relu, Regularization, Dropout, Batch Normalization
Word2vec - Concept and Examples
Introduction and Text Classification Example using Word2vec
Week 4
Convolutional Neural Networks (cnn)
Structure, Layers, Training, Transfer learning, Text Classification exercise
Introduction to Recurrent Networks
Introduction to key concepts and advantages
Week 5
Rnn/lstm deep dive
Challenges with rnn, lstm concepts, exercises
Advance Topics
Attention layers and Applications, End Term project

Jahnavi Mahanta
Jahnavi has 13+years of marketing/risk analytics strategy experience in financial/credit card domain with organizations like American Express. Jahnavi has led many data-science teams to develop and deploy large scale marketing data capabilities like customer offer recommenders, customer response prediction models. A machine learning enthusiast and learner, with a special interest in the area of deep learning.
Manu Chandra
Manu has 15+ years of experience in Analytics and consulting working with leading global organizations like American Express, Accenture in the areas of Fraud Management, Credit risk and Customer analytics. Manu is the co-founder and Chief Data Scientist at MathLogic, an analytics/data science consulting and training firm.