Machine Learning & Deep Learning  Apply into practice
1. Introduction
Machine learning is so popular and widen use today that you probably use it dozens of times a day without knowing it.
For example, searching on the web, speech recognition, translation, selfdriving cars, etc. This course will help you to review math knowledges relate to Machine Learning, algorithms and the most effective of machine learning techniques, and gain practice implementing them in the real case.
More importantly, you will learn not only the theoretical underpinnings of learning, but also gain practical and best practices in projects about Machine Learning/Deep Learning.
2. Training Targets
Share knowledge and share experience from R&D and projects of ML/DL
3. Knowledge and Skills gained after the course
Gain knowledge of algorithms and apply to the real business case to support for business, help them make the right solutions and quickly to reduce cost in the competition environment.
4. Career Opportunities
Machine learning and AI is a big trend today and in the future so you will have more opportunities to work with the companies, projects about ML.
5. Training Period
 Training time is 60 hours
 The Schedule is Saturday  Sunday, from 09:00 to 12:00
 Fee: 9.000.000 VNĐ(Student Fees: 7.500.000 VNĐ)
6. Target
You want to work in Big Data, AI, IoT Technology, R&D and projects of ML/DL
7. Requirements
 Have knowledge about Python
 Linear algebra
 Probability & Mathematic statistics is an advantage.
8. Content
Introduction AI & Machine Learning 
 The concept of AI, Machine Learning
 Key terminology
 Examle to explain norm in ML
 Simple Workflow
 Steps in developing a Machine Learning application

Jupyter Notebook/Python 
 Introduction
 Install or using docker
 using to implement algorithms, virtualize data, plot data
 Practice

Feature Engineering 
 Introduction
 General model
 Feature Extractor
 Feature selection
 Dimensionality reduction
 Feature Scaling and Normalization
 Practice

Softmax Regression 
 Introduction
 Onevsrest model
 Softmax function
 Softmax with python
 Exampe with Simulated data
 Practice

Overfit/Underfit 
 Introduction
 Validation
 Regularization
 Other methods
 Practice

Gradient Descent Section 
 Introduction
 Gradient Descent for function 1 variable
 Example
 Learning rate
 Gradient Descent with multiple variants
 Level set
 Speed of conver by algorithms
 Momentum
 Nesterov accelerated gradient
 Batch Gradient Descent
 Stochastic Gradient Descent (SGD)
 Example with Linear Regression

Classifying with kNearest Neighbors 
 Explain Algorithm
 General approach to kNN
 Handwriting Recognition System
 Demo

Linear Regression 
 Introduction & Defines
 Predicted error
 Lost function
 Find optimal point
 Problems can be solved by Linear Regression
 Practice

Kmeans Clustering 
 Mathematics
 Lost Function
 Algorithm to optimize Lost Function
 Demo with MNIST data  handwriter
 Practice 
Decision Trees 
 Introducing decision trees
 Measuring consistency in a dataset
 Using recursion to construct a decision tree
 Plotting trees in Matplotlib
 Practice

Logistic Regression 
 Introducing
 Example
 Model Logistic Regression
 Lost function
 Find optimal point
 Example with python
 Practice

Binary Classifiers Problems 
 Introduction
 Binary Classifiers for Multiclass Classification problems
 Onevsone
 Hierarchical
 Binary coding
 onevsrest and onehot coding
 Neural Networks Form
 Gender problem based on face image
 Practice

Naive Bayes Classifier 
 Introduction
 Probability review for Machine Learning
 Multinomial Naive Bayes
 Examples
 Practice

Recommendation 
 Introduction Asociation Rules/Apriori/FPGrowth
 ContentBased Filtering
 Useruser collaborative Filtering
 Itemitem Collaborative Filtering
 Practice

Peceptron Learning 
 Introduction
 Problem
 Perceptron
 Example
 Practice

Multilayer Perceptron 
 Introduction
 PLA for basic logic functions
 Layers
 Units
 Weights and Biases
 Activation functions
 Backpropagation
 Example
 Practice

Introduce Deep Learning 
 Introduction
 Concept and workflow
 Base algorithms
 CNN algorithm explain & Approach
 Practice

RNN & LSTM 
 Explain Algorithm
 General approach to RNN
 Practice

Azure Machine Learning 
 Introduce Azure ML
 Components on Azure ML
 How to train model on Azure ML
 How to consume and retrain model on Azure ML
 Practice

Exercise 
 Final Exercise 

 9. Materials
 10. Study Method
 100% study directly on computer, each student studies on a computer.
 2 sessions of theory & 1 practice session in alternating order (Lab exercises / Activities)
 11. Evaluation
 Examination form: theory (multiple choice) and practice (on computer)
http://exam.b4usolution.com/
Satisfactory score is greater than or equal to 45%
 12. After completing, you can:
 Applied Scientist, Robotics and Machine Learning


Register:
Phone: 098 921 4285
Email: info@b4usolution.com  hoalethibich85@gmail.com
Skype: hoa.lethibich






















