Human Compatible: Artificial Intelligence and the Problem of Control
In the popular imagination, superhuman artificial intelligence is an approaching tidal wave that threatens not just jobs and human relationships, but civilization itself. Conflict between humans and machines is see ...more
More lists with this book...
I'll back up a little. Since the beginning o ...more
We must assure good morals,
Inside future A-G-I.
To prevent a future quarrel,
In which many of us die.
And so that we may proceed,
In our ...more
“He who controls the algorithms controls the universe.”
To get just an inkling of the fire we’re playing with, consider how content-selection algorithms function on social media. They aren’t particularly intelligent, but they are in a position to affect the entire world because they directly influence billions of people. Typically, such algorithms are designed to maximize click-through, that is, the probability that the us ...more
The thesis of this book is that we need to change the way we develop AI if we want it to remain beneficial to us in the future. Russell discusses a different kind of machine learning approach to help solve the problem.
The idea is to use something called Inverse Reinforcement Learning. It basically means having AI learn our preferences and goals by observing us. This is in contrast to us specifying goals for the AI, a mainstream practice that he refers to as the “standard model”. Add some game th...more
I mean, we're not there yet, or to be precise, we aren't at the point of AI super-intelligence.
But that doesn't speak to the issue that has gotten a lot of traction in popular media, from movies to scien ...more
It's a book about REAL AI (not statistics!) w/o buzzwords.
These are mainly philosophical considerations (about conscience, instincts, control mechanisms, ethics, superiority and many more) that DO have a lot of practical applicability. What I appreciate is:
* the book doesn't look for cheap publicity ("we're all doomed!")
* it doesn't try to "ride on the hype wave"
* it's really thorough when it comes to different dilemmas - possibly TOO muc ...more
In order to control such an AI, and align it with human interests, Russell propose ...more
Russell somehow manages to communicate the minority view on the importance of the safety / control problem, while remaining grounded in practical problems and research methods. It's such a difficult topic to write a ...more
The primary value I got out of it is his 3-point approach towards solving the problem. I find his argument convincing and will definitely keep these points (see notes below) in mind in the future.
Note: I only documented the latter part of the book; the intro was also very i ...more
A.I.s are in some specific cases already better than humans, e.g. by winning ever more complex games like chess,Go, or StarCraft against human champions. AGIs are not restricted to a specific field but can do intellectual tasks like humans. In the case of strong AGIs, they are even conscious and self-aware.
As soon as a superintelligence gets better than a human, it gets better at e ...more
I did not finish it. I made it ~halfway through and then pulled the plug, which is something I almost never do.
I was excited to read this one, as I am very interested in AI. Author Stuart Russell's delivery left much to be desired, however...
The book is written in an extremely dry and long-winded manner. I found my attention wandering, and was getting irritated. The reading was extremely tedious and jumped around quite a lot.
The final straw was hearing him talk about how c ...more
Russell steps back from this dire prognostication to begin the book with a review of AI, how it has come along to where we are today. The Baldwin effect ...more
Russell calls all of goaldirectedness "the standard model" and says we should declare today year zero of real ai research-- he proposes a nonstandard model that is just three principles. 1. The machine's only objective ought to be satisfaction of human preferences. 2. The machine is uncertain about what those prefer ...more
The book thought me that AI is not to be taken lightly. Too much is at stakes, and nobody knows when the big AI boom will happen. It will probably be the next big catastrophe in our history or the machines might be our saviors, who knows!
All the relevant questions are answered in this book with the best infor ...more
It is actually a 5* book but wasn't overly happy with the style of writing. ...more
AI Pioneer Stuart Russell is a Professor of Computer Science at UC Berkeley, has numerous awards, fellowships, chairmanships, etc. and has co-authored a textbook on AI with Peter Norvig. This is a book written by that rar ...more
Russel likens us to a person asking for wishes from a genie when it comes to developing these superintelligent machines. The third wish is always to undo the first two. We have t ...more
First Law — “A robot may not injure a human being or, through inaction, allow a human being to come to harm.”
Second Law — “A robot must obey orders given to it by human beings except where such orders would conflict with the First Law.”
Third Law — “A robot must protect its own existence, as long as such protection does not conflict with the First or Second Law.”
However, even the most illiterate tech novices can ...more
Russell S (2019) (11:38) Human Compatible - Artificial Intelligence and the Problem of Control
• Why This Book? Why Now?
• Overview of the Book
01. If We Succeed
• How Did We Get Here?
• What Happens Next?
• What Went Wrong?
• Can We Fix It?
02. Intelligence in Humans and Machines
• • Evolutionary origins
• • Evolutionary accelerator
• • Rationality for one
• • Rationality for two
• • The limits of computation
• Intelligent Computers
• • Ag ...more
Given our present trajectory of civilization - in where we are becoming not only more automated, but more reliant on computers, it seems likely that A.I will become omnipresent. This is, in other words, software that can take in new data, react, and change its data structure to respond more approp ...more
First, it is a super-well-informed author who provides a very-well balanced discussion of the issues humans might face upon the emergence of a super-intelligent AI. I'm much more able to get my arms around Russell's arguments than I am Nick Bostrom's Superintelligence: Paths, Dangers, Strategies; though Bostrom is well-informed, his tendency is to inflame fear, while Russell is better at simply making a well-reasoned cas ...more