Simply Artificial Intelligence
Aug. 17th, 2023 10:56 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
quippeThe Blurb On The Back:
Simply - Artificial Intelligence
Sometimes less really is more. This smart but powerful little guide unlocks the inner workings of artificial intelligence, from the first computer programs to the internet of things.
Part of Dorling Kindersley Limited’s SIMPLY SERIES, this is a useful book for anyone interested in understanding the basics of artificial intelligence (“AI”). I found the sections looking at AI’s history to be the most interesting as once it gets into the science I found the explanations pitched at a level a little too high for me to follow while the chapters looking at problems with AI (including societal issues) were disappointingly brief.
The book is divided into 6 sections:
- History of Artificial Intelligence;
- Classical Artificial Intelligence;
- Statistical Artificial Intelligence;
- Using Artificial Intelligence;
- Philosophy of Artificial Intelligence; and
- Living With Artificial Intelligence.
Each section is in turn divided into short chapters of just a couple of pages, each of which deals with a particular idea.
I very much enjoyed the section on the history of artificial intelligence, which explains what computers are and the early days of artificial intelligence. I also found the Classical Artificial Intelligence section pretty easy to follow at first as it sets out the basis for programming, modelling and dealing with uncertainty as it breaks the ideas down into manageable chunks and explains them in ways that I found easy to follow.
Where I started to lose track with the book was with Statistical Artificial Intelligence where the statistics and maths principles being a little too advanced for me, particularly when it came to network structures and processing. To be honest, I didn’t follow an awful lot of it but I think if you have more of a maths/statistics background then it should be pretty straightforward - I just lacked the grounding (which is a shame given that the purpose of this book is to explain things - I think it just needed to break it down a further level).
The Using Artificial Intelligence section is fairly generic and although it gives a comprehensive summary of the various potential uses of AI but it’s quite generic and more disappointingly, uncritical. There’s no consideration here of potential issues with AI, e.g. incorrect datasets, if there is a risk in people putting unquestioning faith in AI information, racial or sexist biases in datasets and the impact it could have on results (although this is considered later in the Living With Artificial Intelligence). Similarly the Philosophy of Artificial Intelligence is fairly brief and focuses on consciousness and the imitation game issue, when it would have been more interesting to me to see a discussion on whether they are capable of creativity (a particular hot topic in recent months).
Living With Artificial Intelligence does consider the impact of bad data (including bias in data) but does so from a generic look when I think bad faith from bad actors needed to be in there because that is a very real danger. Similarly the book looks at the echo chamber effect and filter bubbles is done in a very neutral way as if this is something that may just happen when it happens because it’s profitable for companies to put certain content in front of people and feed them more. Given the media hype about the AI threat, I found the chapters looking at this to be such a disappointment - incredibly basic without any real consideration of what the threat is and how credible it is.
Ultimately, if you’re looking for a starting point into the subject then this works but I don’t think it gets the balance right in terms of the level its audience is starting at and I don’t think it makes any serious attempt to grapple with the issues and challenges that AI throws up.
The Verdict:
Part of Dorling Kindersley Limited’s SIMPLY SERIES, this is a useful book for anyone interested in understanding the basics of artificial intelligence (“AI”). I found the sections looking at AI’s history to be the most interesting as once it gets into the science I found the explanations pitched at a level a little too high for me to follow while the chapters looking at problems with AI (including societal issues) were disappointingly brief.
SIMPLY - ARTIFICIAL INTELLIGENCE was released in the United Kingdom on 2nd March 2023. Thanks to the Amazon Vine Programme for the review copy of this book.
Sometimes less really is more. This smart but powerful little guide unlocks the inner workings of artificial intelligence, from the first computer programs to the internet of things.
Part of Dorling Kindersley Limited’s SIMPLY SERIES, this is a useful book for anyone interested in understanding the basics of artificial intelligence (“AI”). I found the sections looking at AI’s history to be the most interesting as once it gets into the science I found the explanations pitched at a level a little too high for me to follow while the chapters looking at problems with AI (including societal issues) were disappointingly brief.
The book is divided into 6 sections:
- History of Artificial Intelligence;
- Classical Artificial Intelligence;
- Statistical Artificial Intelligence;
- Using Artificial Intelligence;
- Philosophy of Artificial Intelligence; and
- Living With Artificial Intelligence.
Each section is in turn divided into short chapters of just a couple of pages, each of which deals with a particular idea.
I very much enjoyed the section on the history of artificial intelligence, which explains what computers are and the early days of artificial intelligence. I also found the Classical Artificial Intelligence section pretty easy to follow at first as it sets out the basis for programming, modelling and dealing with uncertainty as it breaks the ideas down into manageable chunks and explains them in ways that I found easy to follow.
Where I started to lose track with the book was with Statistical Artificial Intelligence where the statistics and maths principles being a little too advanced for me, particularly when it came to network structures and processing. To be honest, I didn’t follow an awful lot of it but I think if you have more of a maths/statistics background then it should be pretty straightforward - I just lacked the grounding (which is a shame given that the purpose of this book is to explain things - I think it just needed to break it down a further level).
The Using Artificial Intelligence section is fairly generic and although it gives a comprehensive summary of the various potential uses of AI but it’s quite generic and more disappointingly, uncritical. There’s no consideration here of potential issues with AI, e.g. incorrect datasets, if there is a risk in people putting unquestioning faith in AI information, racial or sexist biases in datasets and the impact it could have on results (although this is considered later in the Living With Artificial Intelligence). Similarly the Philosophy of Artificial Intelligence is fairly brief and focuses on consciousness and the imitation game issue, when it would have been more interesting to me to see a discussion on whether they are capable of creativity (a particular hot topic in recent months).
Living With Artificial Intelligence does consider the impact of bad data (including bias in data) but does so from a generic look when I think bad faith from bad actors needed to be in there because that is a very real danger. Similarly the book looks at the echo chamber effect and filter bubbles is done in a very neutral way as if this is something that may just happen when it happens because it’s profitable for companies to put certain content in front of people and feed them more. Given the media hype about the AI threat, I found the chapters looking at this to be such a disappointment - incredibly basic without any real consideration of what the threat is and how credible it is.
Ultimately, if you’re looking for a starting point into the subject then this works but I don’t think it gets the balance right in terms of the level its audience is starting at and I don’t think it makes any serious attempt to grapple with the issues and challenges that AI throws up.
The Verdict:
Part of Dorling Kindersley Limited’s SIMPLY SERIES, this is a useful book for anyone interested in understanding the basics of artificial intelligence (“AI”). I found the sections looking at AI’s history to be the most interesting as once it gets into the science I found the explanations pitched at a level a little too high for me to follow while the chapters looking at problems with AI (including societal issues) were disappointingly brief.
SIMPLY - ARTIFICIAL INTELLIGENCE was released in the United Kingdom on 2nd March 2023. Thanks to the Amazon Vine Programme for the review copy of this book.
