2-days Instructor-led
COURSE OVERVIEW
This introduction-level hands-on course explores the field of artificial intelligence (AI) security, it’s essential to identify and address various types of attacks and risks that can compromise the integrity and reliability of AI models and systems. Data Security Risks – The AI Pipeline as an Attack Surface: The whole data pipeline is vulnerable to assaults since AI systems rely on data.
In this course, we’ll cut through the math and you’ll learn exactly how machine learning algorithms work. We’ll focus on the security implications using clear explanations, simple Python code (no libraries), and step-by-step labs, you’ll discover how to load and prepare data, evaluate your models, and implement a suite of linear and nonlinear algorithms along with assembling algorithms from scratch.
This course presents a wide variety of related technologies, concepts, and skills in a fast-paced, hands-on format. This provides you with a solid foundation for understanding and getting a jumpstart into working with AI from a security and risk-analysis perspective.
WHAT YOU’LL LEARN
Join an engaging hands-on learning environment, where you’ll explore:
- Getting Started with Python and Jupyter
- Statistics and Probability Refresher and Python Practice
- Matplotlib and Advanced Probability Concepts
- Algorithm Overview
- Predictive Models
- Applied Machine Learning
- Recommender Systems
- Dealing with Data in the Real World
- Machine Learning on Big Data (with Apache Spark)
- Testing and Experimental Design
- GUIs and REST: Build a UI and REST API for your Models
- Challenges to Risk Management
- Tracking Emergent Risk
- Availability of Reliable Metrics & Risks at Different Lifecycle Stages
- Measuring Risk in Real World Settings
- Challenges to Risk Tolerance & Prioritization
- Organizational Integration & Management of Risk
- How AI Risks Differ from Traditional Risks
Virtual Classroom Live Outline
Getting Started with Data in Machine Learning World
- Training Data: The part of data we use to train our model. This is the data that your model actually sees(both input and output) and learns from.
- Validation Data: The part of data that is used to do a frequent evaluation of the model, fit on the training dataset along with improving involved hyperparameters (initially set parameters before the model begins learning). This data plays its part when the model is actually training.
- Testing Data: Once our model is completely trained, testing data provides an unbiased evaluation. When we feed in the inputs of Testing data, our model will predict some values(without seeing actual output).
- After prediction, we evaluate our model by comparing it with the actual output present in the testing data. This is how we evaluate and see how much our model has learned from the experiences feed in as training data, set at the time of training.
Statistics and Probability Refresher and Python Practice
- Types of data
- Mean, median, and mode
- Using mean, median, and mode in Python
- Standard deviation and variance
- Probability density function and probability mass function
- Types of data distributions
- Percentiles and moments
Matplotlib and Advanced Probability Concepts
- A crash course in Matplotlib
- Covariance and correlation
- Conditional probability
- Bayes’ theorem
Algorithm Overview
- Data Prep
- Linear Algorithms
- Non-Linear Algorithms
- Ensembles
- Predictive Models
- Linear regression
- Polynomial regression
- Multivariate regression and predicting car prices
- Multi-level models
Applied Machine Learning with Python
- Machine learning and train/test
- Using train/test to prevent overfitting of a polynomial regression
- Bayesian methods: Concepts
- Implementing a spam classifier with Naïve Bayes
- K-Means clustering