Hybrid Symbolic AI and Deep Learning Techniques and Applications

I have been working in the field of artificial intelligence since the 1980s and starting in 2014 have taken a deep dive into deep learning. This book provides complete practical examples for intelligent systems applications. Think of this book as being a "cook book," but with long recipes for:

Working with Data

Deep Learning

Natural Language Processing

Bayesian Optimization

Reinforcement Learning and Multi-Arm Bandits

Data Storage with Relational and Graph databases

Semantic Web and Linked Data

Python now has the largest share of AI developers' attention and use. In the past I have used the following languages for both my day to day work and for material for books I have written: Common Lisp, Scheme, Prolog, Java, and Ruby. In recent work (manager of a machine learning team) and for most of my consulting work in recent years, I have used Python. I use languages and tools that offer me the most advantages for getting my work done.

Here you will learn how to use what I consider to be the best Python libraries for building intelligent systems. You will work through examples using NumPy, Tensorflow, Keras, fasttext, PyMzn (Python wrapper for the MiniZinc discrete optimization language and family of solvers), ProbLog (for Probabalistic Logic), and Pomegranate (for Probabalistic Graph Models). You will use SqLite, Postgres, Neo4J and RDF data stores.

My goal was to write a book for you that you would have fun with while working through the examples and to provide a source you can use later when you have specific project requirements.