DeapSECURE module 4: Deep Learning (Neural Network)

In this lesson module, we will introduce deep learning using neural network models, which is very powerful in addressing some pressing challenges in the area of cybersecurity.

The objectives of this deep learning lesson are to help learners to:

Understand the needs of large-scale machine learning in cybersecurity.
Understand the fundamentals of neural networks and their properties.
Gain practical familiarity with neural network model architecture and construction.
Gain practical experience with neural network training & inference.
Improve (tune) neural network models.
Use the computing power of HPC to effectively train, tune, and deploy neural network models.

Prerequisites

Knowledge and experience of Python programming language. Please refer to DeapSECURE’s list of crash courses on Python for pointers to useful resources to get started with Python programming.

Basic understanding of data analytics and machine learning. We recommend that learners have taken DeapSECURE’s Big Data and Machine Learning lessons to equip themselves with the fundamental knowledge and experiences needed in this lesson on neural networks.

Website: https://deapsecure.gitlab.io/deapsecure-lesson04-nn

Schedule

	Setup	Download files required for the lesson
00:00	1. Introduction to Machine Learning & Deep Learning	What is deep learning? What are the top deep learning algorithms? What are the differences between Artificial Intelligence (AI) vs Machine Learning (ML) and Deep Learning (DL)?
00:00	2. Deep Learning to Identify Smartphone Applications	What are the security challenges related to smartphones? What approaches are being used by security researchers to devise better ways to secure mobile devices? How can large datasets like SherLock help researchers detect security issues on smartphones? How can the SherLock dataset be used to develop a method for identifying smartphone apps?
00:25	3. Overview of Deep Neural Network Concepts	What is a neuron? How does a neuron work? What is a neural network? How is a neural network trained?
00:25	4. An Introduction to Keras with Binary Classification Task	What is Keras? What are the advantages of using Keras to build neural network models? What are the basic steps for building a neural network model in Keras?
00:25	5. Classifying Smartphone Apps with Keras	How do we build a neural network on KERAS to perform multi-class classification? How do we monitor the progress of neural-network training? How do we set up appropriate hyperparameters?
01:25	6. Tuning Neural Network Models for Better Accuracy	What is model tuning in Deep Learning? What are the different types of tuning applicable to a neural network model? What are the effects of tuning a particular hyperparameter to the performance of a model? Is Jupyter notebook the best platform for such experiments?
02:30	7. Effective Deep Learning Workflow on HPC	How do we train and tune deep learning models effectively using HPC? How to convert a Jupyter notebook to a Python script? How can we perform post-analysis of HPC computations using Jupyter?
03:35	8. Post-Analysis for Modeling Tuning Experiments	How should we visualize the outputs from the post-processing phase? What general trends do we observe from the last epoch results?
04:40	9. Using Multicore CPUs or GPUs for KERAS Computation	How do we use multicore CPUs for KERAS computations? How do we use GPUs for KERAS computations?
04:40	10. Dealing with Issues in Data and Model Training	What are common data issues in Machine Learning (ML)? How do data issues impact ML model performance? How to identify data issue effects on ML model performance? How to address these data issues? How to identify issues in ML model training caused by overfitting and underfitting? How to address overfitting and underfitting?
05:40	11. Deep Learning in the Real World	What kinds of layer can be built into a neural network? Based on different questions, what kind of layers should we used?
05:40	Finish

The actual schedule may vary slightly depending on the topics and exercises chosen by the instructor.