Overview

This hands-on introductory training in machine learning allows participants to understand and program a range of models for supervised, unsupervised and reinforcement learning. We use a number of business case studies to apply machine learning to, and help participants identify use cases from their own professional environment. The practical part of the training uses Python libraries.

Target Audience

This course is designed for IT or business professionals who seek a practical and thorough introduction to machine learning and data science.

Objectives

• Understand the fundamental concepts of AI, machine learning and data science.
• Discover the principal machine learning models and be able to identify the model best suited to each business case.
• Become familiarized with the important Python libraries for machine learning.

Duration

2 or 3 days.

Program

• Introduction and Key Concepts
• Supervised Learning, unsupervised learning, reinforcement learning, deep learning, classification v regression.
• Exercise: introducing basic Python libraries (Numpy, Matplotlib, Pandas, ...)
• Supervised Learning
• Models
• k-NN, linear and polynomial regression, logistic regression, Guassian Naive Bayes, ...
• Deep Learning - Artificial Neural Networks
• Exercise on use cases: predictive maintenance, customer churn, ...
• Machine Learning Projects
• Data cleaning, normalization, value encodings, ...
• Feature reduction: principal component analysis, linear discriminant analysis
• Model over-fitting v under-fitting
• Risks
• Unsupervised Learning
• Models
• Clustering (k-means), association rules
• Exercise on use cases: product recommendations, customer insights, ...
• Reinforcement Learning
• Models
• Thompson Sampling, Q-learning, deep Q-learning, ...
• Exercise: use cases for each model
• Model Validation
• K-fold validation, GridSearch.

Contact

Contact us for more details.