Demystifying Artificial intelligence
US$ 19.95
The publisher has enabled DRM protection, which means that you need to use the BookFusion iOS, Android or Web app to read this eBook. This eBook cannot be used outside of the BookFusion platform.
Description
Contents
Reviews
Language
English
ISBN
9789389898705
Cover Page
Title Page
Copyright Page
Dedication Page
About the Author
About the Reviewers
Acknowledgement
Preface
Errata
Table of Contents
1. Introduction
Structure
Objectives
What is data?
1. Text
2. Images
3. Audio
What is artificial intelligence (AI)?
Types of artificial intelligence
1. Narrow AI
2. General AI
What is machine learning (ML)?
Conversation time
1. Supervised machine learning
2. Unsupervised machine learning
What is deep learning?
What’s a neuron then?
What is “deep” in deep learning?
Advantages of deep learning
Limitations of deep learning
What is data science?
Why do we need artificial intelligence (AI)?
Automation
Conclusion
Questions
2. Going Deeper into ML Concepts
Structure
Objectives
Machine learning and Maths
Why is ML dependent on Maths?
Types of machine learning
1. Supervised machine learning
Conversation time
2. Unsupervised machine learning
Conversation time
Problem types in supervised learning
1. Classification type problems
Conversation time
2. Regression type problems
Conversation time
Problem types in unsupervised learning
1. Clustering
2. Dimensionality reduction
Semi-supervised learning
Self-supervised learning
How does a machine “learn” to perform a task?
1. Overfitting
Conversation time
2. Under-fitting
How does a machine “learn” to do things?
What is loss and loss function?
Conversation time
How does the neural network “learn” to perform a task?
What are “weights” in neural networks?
Why do we need a “deep” network or multiple layers in a network?
What is reinforcement learning (RL)?
What is the “reward” in RL?
Conversation time
Real-world examples
1. Gaming
2. Chemistry reactions
What is transfer learning?
Why do we do it?
Transfer learning in deep learning
Conversation time
What is computer vision?
1. Image classification
2. Image detection
3. Image segmentation
Downsides/flaws of computer vision
1. Adversarial attack
What is natural language processing (NLP)?
Attention mechanism in NLP
1. Text cleaning and preprocessing
2. Building ML models
Genetic algorithms in ML
1. Population
2. Fitness calculation
3. Parent selection
4. Crossover
5. Mutation
6. Offspring
Real-world applications
1. Medical and health care
2. Vehicle routing problems
Generative adversarial networks (GANs)
Conversation time
Recommendation
How do recommendations work?
Conclusion
Questions
3. Business Perspective of AI
Structure
Objectives
How do AI/ML projects work in the real world?
1. Deciding the task
2. Collecting data
3. Pre-processing data
What is feature engineering?
4. Training the computer/model
5. Checking whether the model has learned what we wanted it to or not
Conversation time
6. Use a trained model for future unseen test data
How to monetize AI?
How can one company/individual monetize AI?
Traditional businesses
Limitations of traditional non-tech businesses
Solutions of problems for traditional businesses
After achieving the growth path, how to maintain it?
How can you leverage AI in your business?
High quantity and quality data
Best talent who knows AI
Domain expertise
How to use AI in your existing products?
Let’s take a look at each of them in detail.
Automating the most time and cost consuming part of the process
Increasing sales with the product using data we already have
Demand forecasting
Dynamic pricing
Adding new AI-backed features in our existing products
Real-world use cases of AI
Netflix recommendation
Self-driving vehicles
How self-driving vehicles work?
Limitations and advantages of machine learning
Advantages
Automation
Speed
Performance
Limitations
One machine can’t do it all (i.e., one model can’t do multiple tasks)
Lake of explainability (also known as trust deficit)
It’s not that hard to fool the machine
Conclusion
Questions
4. How to Get Started, and Pitfalls to Avoid in AI
Structure
Objectives
How to get started in AI and machine learning?
How to start?
How to grow?
How to maintain growth?
A realistic view of artificial intelligence (AI)
Artificial intelligence and employment
Loss of jobs
1. Self-driving vehicles
2. Transportation
3. Education
4. Customer service/experience
5. Defense and security
Pitfalls to avoid in artificial intelligence
1. AI is not magic
2. Performance of AI models will degrade over time
3. Biases in our data can sometimes cause serious problems
4. Human help is still required
5. AI field is at a very early stage
Conclusion
Questions
Quiz time!
Answers for the quiz
References
Index
Loading...