engineering: 2015

What are the different tracks that can end in, "A.I. gain[ing] the ability to improve itself, and in short order exceeds the intellectual potential of the human brain."
http://www.newyorker.com/magazine/2015/11/23/doomsday-invention-artificial-intelligence-nick-bostrom
The people who say that artificial intelligence is not a problem tend to work in artificial intelligence. [...] Oren Etzioni, the C.E.O. of the Allen Institute for Artificial Intelligence, in Seattle, referred to the fear of machine intelligence as a “Frankenstein complex.” Another leading researcher declared, “I don’t worry about that for the same reason I don’t worry about overpopulation on Mars.” Jaron Lanier, a Microsoft researcher and tech commentator, told me that even framing the differing views as a debate was a mistake. “This is not an honest conversation,” he said. “People think it is about technology, but it is really about religion, people turning to metaphysics to cope with the human condition. They have a way of dramatizing their beliefs with an end-of-days scenario—and one does not want to criticize other people’s religions.” [A digit-al god.]

Eventually, the researchers started to question the goal of building a mind altogether. Why not try instead to divide the problem into pieces? They began to limit their interests to specific cognitive functions: vision, say, or speech.

In the history of computer science, no programmer has created code that can substantially improve itself. [But it doesn't need to. Any minute measure could be replicated, improved. Unless Khatchadourian (the author of this) means that no program has been created that does anything related to 'intelligence.' Can a program edit itself? Do they merely boil down to 1s & 0s, nothing more? An example, unrelated to the last strand of thought, comes to me as voice recognition software. I'm told that it can adjust to your tendencies, but this is not intelligence in the self-autonomous realm. That is just taking a multiple of different programs and cherry picking; starting at a sub-root of the tree, {{cont. taking in reverse order}} jumping onto different limbs, then different branches, multiple trunks, stems, sub-stems, and finally (at computer computational speed), the preferred route. Just nodes on nodes.]

The book begins with an “unfinished” fable about a flock of sparrows that decide to raise an owl to protect and advise them. They go looking for an owl egg to steal and bring back to their tree, but, because they believe their search will be so difficult, they postpone studying how to domesticate owls until they succeed. Bostrom concludes, “It is not known how the story ends.”

“Artificial intelligence already outperforms human intelligence in many domains.” The examples range from chess to Scrabble.

One program from 1981, called Eurisko, was designed to teach itself a naval role-playing game. After playing ten thousand matches, it arrived at a morally grotesque strategy: to field thousands of small, immobile ships, the vast majority of which were intended as cannon fodder. In a national tournament, Eurisko demolished its human opponents, who insisted that the game’s rules be changed. The following year, Eurisko won again—by forcing its damaged ships to sink themselves.

Given even the most benign objective—to win a game—such a system, Bostrom argues, might develop “instrumental goals”: gather resources, or invent technology [...].

The brain of the village idiot and the brain of a scientific genius are almost identical.

wtf is this saying:
A respected minority of A.I. researchers began to wonder: If increasingly powerful hardware could facilitate the deep-learning revolution, would it make other long-shelved A.I. principles viable? “Suppose the brain is just a million different evolutionarily developed hacks:

[...] deep learning. Perhaps the most interesting acquisition is a British company called DeepMind, started in 2011 to build a general artificial intelligence. Its founders had made an early bet on deep learning, and sought to combine it with other A.I. mechanisms in a cohesive architecture. In 2013, they published the results of a test in which their system played seven classic Atari games, with no instruction other than to improve its score. For many people in A.I., the importance of the results was immediately evident. I.B.M.’s chess program had defeated Garry Kasparov, but it could not beat a three-year-old at tic-tac-toe. In six games, DeepMind’s system outperformed all previous algorithms; in three it was superhuman. In a boxing game, it learned to pin down its opponent and subdue him with a barrage of punches.

DeepMind’s system still fails hopelessly at tasks that require long-range planning, knowledge about the world, or the ability to defer rewards

Read the section before coming to class, it would help. In class, ask lots of Qs.
Keep on trying with hard problems: the answer will come sooner or later. Don't give up. Eventually the answers will come to you; if not, ask. Expect that you will get stuck on problems and need extra help.
Maintain a list of hard problems in a separate or in the back of the notebook. Practice doing these problems over and over until you can write down the complete correct solution without any help.
Strike middle ground between conceptualizing and practicing. They both draw upon each other to efficiency.
Review by explaining challenging types of problems. (1) Get a blank sheet of paper (2) Explain the technique or concept from sheer recollection.

General pointers