Did you Miss the Kickstarter Project?
The Kickstarter project ended on March 22, 2014. If you would like to see the current status of this book series please click here!
I need your support to continue a series of books explaining Artificial Intelligence (AI) in a mathematically gentle way. The series is called "Artificial Intelligence for Humans". The book that I am proposing for this Kickstarter campaign is Volume 2: Nature Inspired Algorithms. The other volumes will likely follow as future Kickstarter projects. This book is the second in a series, but can easily be read independently.
This book focuses on Artificial Intelligence algorithms that were inspired by nature. This includes genetic algorithms, genetic programming, ant colony optimization, swarm intelligence, cellular automata and artificial life. None of these algorithms emulate every aspect of nature. That would be far too complex. Rather, we take inspiration from many different aspects of nature. Not everything nature creates with carbon and water works as will with human engineered silicon based logic.
This book is targeted at the typical computer programmer who knows at least one computer language and has knowledge of algebra. The book does get into more complex mathematical topics, but I do not assume any knowledge beyond algebra. This worked well for my previous books. If you do a Amazon Search on "neural networks" one of my books is usually first or second. (at least as of this writing)
Examples will be provided in the following languages:
If funding goals are met, I will add stretch goals for additional programming languages. Languages such as C, C++, R, Scala and F# are all candidates. I will also look to the community for help with translation.
This is the second book in a series featuring Artificial Intelligence. The first volume was funded by a previous successful Kickstarter campaign.
Why nature? As of this point, nature is the only one to actually crack Strong Artificial Intelligence (AI). Strong AI is a hypothetical artificial intelligence that matches or exceeds human intelligence.
This book will be published in both paperback and ebook format.
The ebook will be released without DRM.
DRM is a process used to lock down copyrighted material and ensure that customers do not pirate media. Unfortunately, DRM usually punishes the customers who have legitimately purchased the material. If you buy this book you will be given normal PDF/MOBI/ePUB files. You can use them on whatever device is capable of displaying them. You will not need to activate, register or unlock the files.
Artificial Intelligence (AI) is a fascinating topic. I truly enjoy programming in this area. I am excited to bring this expanded book series to market. Wouldn't you like to be a part of this as well? Your pledge will help make this a reality. This will help pay for editing, cover design and other expenses.
I would like to expand the editing for this book. This does cost money. Ideally I use an editor, technical editors and a final proof reader. Self publishing allows me to correct errors even after publication. However, I really do prefer to have the best possible product, as early as possible. You may have even found a typo in this Kickstarter text. This is what I want to try to avoid in volume 2 of this series.
Help me spread the word. If you know of someone who might be interested in this project, please pass the word on. If you have any questions, please post a comment.
Most AI books assume you just walked off a university campus having aced Calculus and Linear Algebra. These books are full of formulas and very short on examples.
I created the Artificial Intelligence for Humans (AIFH) series to make AI more accessible. The volumes planned in this series are as follows.
Only volume 2 is covered by this Kickstarter project. Volume 1 was covered by a previous Kickstarter project. It is not essential that you read Volume 1 first. If you want Volume 1, there are backer levels for that. Volume 0 is a "planned prequel" that will be written after Volume 5 is done.
This book will include chapters covering many "nature inspired algorithms". The tentative table of contents is listed here.
- Chapter 1: Population and Scoring
- Chapter 2: Crossover and Mutation
- Chapter 3: Genetic Algorithms
- Chapter 4: Modeling Problems
- Chapter 5: Genetic Programming
- Chapter 6: Bayesian GP
- Chapter 7: Speciation
- Chapter 8: Particle Swarm Optimization
- Chapter 9: Ant Colony Optimization
- Chapter 10: Artificial Life
Each chapter also includes at least one programming example. Some chapters include multiple examples. The screen shots below show some of these examples. More examples will be added as I write the book. Obviously, not all of the examples are complete yet.
The book starts with Genetic Algorithms. You will see how to evolve solutions to solve a problem. Mother and Father solutions can produce a child solution that shares traits of each.
Genetic programming can be applied to the Traveling Salesman Problem (TSP). This problem seeks to find the shortest route to visit a series of cities. Here you see a solution to the TSP evolving.
You can also evolve programs and equations. Here you see two equations produce a child equation.
Particles are another nature inspired algorithm. Flocks of particles can be used to solve problems.
You can also evolve Artificial Life.
These are only the examples that I have completed so far. The finished book will include many more examples.
The book itself is "language agnostic". The book presents the topics with text, diagrams, pseudocode and formulas. Each chapter tackles one specific topic. Programming examples are provided for each chapter. These examples are all open source (Apache 2 license) and available on GitHub. Keeping the text language agnostic allows me to easily add programming language without the need to revise the series.
Why learn AI? AI is Coming of Age! (seriously). AI does have a major "coolness factor". Many people jump into AI for this reason alone. However, in recent years AI is becoming more important.
Why is AI important? Seriously, AI is coming of age this time! I know AI has been "coming of age" since the 1950's.
The amount of computer data in the world increasing at an amazing rate. This is thrusting many AI related concepts into the mainstream. How many of these terms have you seen in the media lately?
Terms such as Business Intelligence, Data Mining, Predictive Modeling, Big Data and Data Science are all, in various degrees, connected to AI. The world has more data that it can wrap its head around!
Cloud computing is also helping to allow Big Data to be analyzed by AI. Amazon EC2 allows you to quickly command a large cluster of computers to process Big Data tasks.
Harvard Business Review even called "Data Scientist" the sexiest job of the 21st century! (link)
There you go! AI can even help your love life!
Some of the reward levels allow you to choose a SWAG item. There are currently five items to choose from. These are all quick mock-ups, real SWAG items might be slightly different.
You can choose between either a coffee mug or mouse matt. Two designs are available -- Encog the Neuron or the DNA Helix. Or you might choose to have your own "Rule 30 Shell". Of course nature has probably not created two of the same Conus Textile shell, so yours will not look exactly like the above specimen, but it will be about that size.
Important information common to all rewards!
- All AIFH series ebooks are available in PDF, ePUB, PDF & MOBI.
- All of my previous ebooks, covered in eBook extravaganza, are in PDF form.
- Some levels include a "SWAG item", for these levels you must choose ONE swag item to be sent with your orders. Additionally you will get a sticker.
- All rewards that involve previously published ebooks (i.e. AIFH Vol1 and ebook extravaganza) will receive access to those books once the Kickstarter campaign ends successfully.
- With the exception of the "sticker only" level, ALL physical shipments will occur together in one shipment. Sticker-ONLY levels will be sent once the Kickstarter campaign ends successfully.
- EBook Extravaganza includes every AI book I've published so far. You can see the list here.
- A personal autograph means I include your name in the autograph (i.e. Thank you James for supporting my Kickstarter project). A regular autograph does not include your name). The autograph is optional for levels that include it. I am only autographing AIFH Vol 2.
- Every reward level that involves shipping, also gets an Encog Sticker.
If you have questions, let me know. If you have suggestions for other levels that I did not think of, also let me know.
I set the book's funding goal to be $2500. The book quickly shot through this. Maybe I should have set it higher? However, $2500 is really, the minimum, of what I need to produce a book. So this is why I set the funding amount to that level. To make the book an even better product, there are stretch goals. The last project hit nearly $19k so it is very possible to hit these.
$7,500 USD - Extended editing/proof reading. Vol 1 had a single round of editing, and allowed a few typos to "escape". I would like an additional round of editing.
$15,000 USD - Additional Capstone project. The book will currently have one capstone project. If this stretch goal is met, there will be a capstone in BOTH Data Science and Artificial Life.
$20,000 USD - Additional programming language. Right you get Java, C# and Python. Another language will be added, and put to vote. Options are: C/C++, R, Scala, F# & PHP. Might take suggestions for others if particularly strong interest.
Break the $20k barrier, and I will come up with more!
Risks and challenges
There are not too many risks. My last Kickstarter was completed on time. If I do encounter any setbacks I will be open and honest about them. If there are problems or challenges, you will get updates!
Additionally... A black-ops government agency unleashing a "SkyNet" scenario, would be a setback for sure. Large "civilization ending" meteors are bad too.Learn about accountability on Kickstarter
Yes, the books are designed to be largely independent of each other. Reading Volume 1 will give you a deeper understanding of a few of the base algorithms used in Volume 2. However, it is not essential to read Volume 1 first.
Small typo on the $9 and $25 pledges. Both say "Vol 1: Nature Inspired Algorithms", should be "Vol 2: Nature Inspired Algorithms". I fixed this by marking the old $9/$25 pledges as "sold out". The new versions are exactly the same, just with the typo fixed. If you are a backer on the old pledge you do not need to move your pledge. If you do move your pledge, it won't hurt, though.
Support this project
- (30 days)