Speaker: Clarence Chio

Introduction to AI/ML for Software Engineers" is a fast-paced learning path on machine learning from a software professional’s point of view. The class is designed with the goal of providing students with a hands-on introduction to machine learning concepts and systems, as well as giving them the practical skills to walk away with the foundational skills to embark on ML projects in a professional setting.

Over the course of two days, attendees will be put through several hands-on exercises that stimulate learning through writing and executing code, instead of passive lectures. Students will get first-hand experience at cleaning data, implementing machine learning programs, and solving real problems in tuning, deploying, scaling, and maintaining machine learning systems.

Each attendee will be provided with a comprehensive virtual machine programming environment that is preconfigured for the tasks in the learning path, as well as any future machine learning experimentation and development that they will do. This environment consists of all of the most essential machine learning libraries and programming environments friendly to even novices at machine learning. As a capstone at the end of the session, students have a chance to formulate and embark on implementing a real machine learning system, from data collection to deployment.
 

Day 1 Topics

1. Introduction to machine learning

• Hands-on guided exploration of Python machine learning libraries:
  • Data-wrangling using Numpy and Pandas
  • Scikit-learn’s functions and capabilities
  • Data visualization using Matplotlib/Seaborn
     

2. Walkthrough of the most commonly used machine learning algorithms (with quick hands-on examples/visualizations for select algorithms)

• Supervised learning algorithms
  • Linear/logistic regression
  • Support Vector Machines
  • Decision trees/Random forests
• Unsupervised learning algorithms
  • Clustering
  • Semi-supervised learning
     

3. Two-hour example: Building (and bypassing) an email spam filter with scikit-learn

• Loading data efficiently
• Using a labelled email/spam corpus training and test set, extract salient features to build a word model of spam
• Model tuning, cross-validation, and evaluation process
• With complete knowledge of the system, manually craft a piece of spam to bypass the filter
   

4. Principles behind selecting the best machine learning models for different use-cases
    
5. Solving practical problems in real-world machine learning deployments

• How to explain the predictions made by your model (using LIME)
• How to approach the problem of class imbalance (using imbalanced-learn)
• How to approach model/result evaluation in an unbiased way
• How to efficiently approach model parameter tuning (grid search etc.)

Day 2 Topics

6. Deep learning

• Using Keras/TensorFlow for anomaly detection with convolutional neural networks
• Choosing the appropriate model for implementing different types of problems: efficacy comparison of different machine learning techniques for solving the anomaly detection problem, and what other considerations to have
   

7. Two-hour example: Building a simple network intrusion detection system with two different machine learning models

• Importance of understanding the data and the threat model before designing a solution for the problem
• Model tuning, cross-validation, and evaluation process
• Guided comparisons of the performance characteristics for each implementation
• Visualizing and presenting the data for ease of analysis by security operation professionals.
   

8. Streaming pipelines for machine learning using Apache Spark MLlib (PySpark)

• Overview of Apache Spark
• General architecture
• Distributed, scalable machine learning deployments with Spark
• Guided example of a streaming architecture for network anomaly detection using reinforcement learning on Spark
   

9. Evaluating machine learning systems 

• Techniques in bias detection, performance/efficacy measurement, and error analysis
• Evaluation of learning system architecture in adversarial scenarios
   

10. Capstone project (in teams of 1-3)

• Formulate a real machine learning system
• Design a strategy for data collection, feature engineering, model selection, deployment, scaling, maintenance, and version control.

Practical Elements

All of the modules from 3-10 (described above in course outline) are practical elements.