Introduction Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

What is meant by a Golem and how is it relevant to this topic?

A

A Golem was a little character made out of clay in Jewish culture in which would be blown the essence of life and it would act autonomously. Humans have always fantasised about creating sentient creatures themselves, it characterises the idea of artificial life. It also reflects how little we knew about how people think: we didn’t even know that we needed a brain to think.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is meant by the uncanny valley?

A

The concept of artificial intelligence often elicits a high arousal response whether positive or negative. Mori’s graph represents the fact that people like robots the more they seem like humans to an extent, until they get too close and they suddenly seem very scary. This dip is called the uncanny valley. Likability increases again as the robots become very like humans.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What was Denny’s suggestion for an explanation for the uncanny valley?

A

We are naturally predisposed to be fearful or aversive to disease and they are too close to people who are ill

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What does the famous video of the abuse of the boston robots demonstrate about our relationship with AI?

A

Human beings have a very strong tendency to anthropomorphize. i.e. they assign human qualities to anything that behaves like a human (and to things that don’t). The video often makes us feel sympathy for them

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is an abacus and explain its relevency

A

A counting frame, used by the sumerians (now Iraq) which ‘outsourced’ mental processes (allowed to compute outside of our head), similar to notebooks. As such we extend our mind and make it more powerful.

This is essentially the beginning out AI and computers, our species has gradually extended its mind to a level that was unimaginable for most of history.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

When/ what was the first idea of a machine being able to do intellectual work?

A

It is a recent fantasy: Mechanical Turk was an 18th century apparatus that could play a strong game of chess. However this was an illusion: the mechanical clockwork inside had a dwarf who was actually playing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Describe when a computor was actually ‘fully fantasised’

A

Charle’s Babbage thought up of a hypothetical analytical engine which would allow the user to specify general calculations in the early 1800’s. He described processors, memory, input etc but it was never built.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

How did Lady Lovelace contribute to this concept?

A

Ada Lovelace was the first person to write a specification to generate bernoulli numbers: an algorithm. These specifications constitute the first software program, and Lady Lovelace is therefore the first programmer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Why was the analytical engine never built?

A

In 1878, the British Association for the Advancement of Science described the engine as “a marvel of mechanical ingenuity”, but recommended against constructing it due to it being ‘esoteric’, not returning any profit or advancement.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

When and how did the idea of rebuilding a human mechanically take hold?

A

The industrial revolution makes the machine mainstream. In factories, machines start to take over human tasks. The idea that one could rebuild a human mechanically takes hold. The idea of a robot (after the Slavic Robota, which means “forced labourer”) is born, however this is not yet linked to the computer

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What invention is the blueprint for every computer we have?

A

The turing machine (Alan Turing, english mathematician who cracks the Enigma code of the Nazis): Turing shows how to make a very generic “machine” to do calculations: the Turing Machine. It was a specification (not a material product) that proves that such a machine can handle all computable functions. This also includes all standard logic: the Turing machine thus can “think” a bit.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Describe some of the specifications of the Turing machine

A

The machine has a pen and can write 0’s or 1’s on a tape, this tape is the storage of the system. It can also read and erase what is on the tape. There is a programme installed which tells the machine what to do when it reads something from the tape.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How is logic implemented into a computer?

A

Through boolean algebra: True sentences are given the value “1”, false sentences the value “0.” In logic, the truth of compositions (“p and q”) is a function of the truth of the elementary statements (“p”, “q”). In Boolean algebra the truth value of compositions is calculated by mathematical operations on the elementary parts. E.g.: the value of “p and q” equals the value of p multiplied by the value of q. “p and q” is thus only true (“1”) if p and q are both true (both “1”), because only then you get 1x1=1(in all other cases you get 0)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What was Denny’s reasoning that the Turing machine can ‘think’?

A

Our brain also can carry out logic through using computational processes which are implemented in the brain. This machine implements a process which is thinking.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What theory did this notion of the thinking Turing machine spark?

A

The computational theory of mind: Thinking consists of computations on representations. These computational processes are realised in the human brain (somehow) just like logic is realised in digital computers. Therefore mental processes are software that runs on the brain’s hardware. (We are programming our computer through this lecture.)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is meant by the Turing test?

A

Turing posited that if we had a computer which we could not distinguish from a human being, then we would have to conclude that it has consciousness. This is called the Turing test.

17
Q

What was the point Turing was trying to make with this hypothetical test?What is meant by the strong AI thesis?

A

In the late 1970s and early 1980s the hopes are high for real robots. The Strong AI thesis states that we will build intelligent, conscious computers. Even though all these building blocks in a computer may not have consciousness, consciousness may emerge as it does in our own brains.

18
Q

Describe the Chinese room criticism of the Turing test

A

A english man given a script in how to respond to certain Chinese letters with other Chinese letters may get better and quicker at the task but will never understand what he is actually doing or saying. This paper argues that executing a computer program in itself is never enough to realise consciousness

19
Q

Distinguish between strong AI and Weak AI theories

A
  • Strong AI = AI will be conscious

* WeakAI = AI is useful to study the mind

20
Q

What is meant by AGI?

A

Artificial General Intelligence (AGI) = AI that is generic in its operation (can do many different tasks- rather than task specific e.g chess). This AI can be presented with a novel game and learn the rules of the game and play it

21
Q

Does strong AI == AGI?

A

Although many people act as if AGI=Strong AI, that need not be the case

22
Q

What two main types of responses were there to the chinese room?

A
  1. There is something in the room (or the room itself) that has a rudimentary form of conscious thought (This was all theorised before we had such functional AI)
  2. The argument shows that the Turing test is not reliable: a system can pass the Turing test without conscious thought
23
Q

Almost all commentaries focus on whether computers actually instantiate cognition, where is the error in this line of reasoning?

A

Where are affect and motivation? In the influential book Descartes’ error, Damasio argued that affect and cognition are not neatly separated. Affect signals what matters to us, and as such prioritises. Without affect, is thinking even possible?

24
Q

What is meant by the AI winter?

A

During the AI winter, nobody believes in AI anymore. Promises had not been kept; in the late 1980s computers could do lots of things, but not much that would be seen as intelligent. However, in a dark corner of science, psychologists like Geoffrey Hinton were developing neural nets…

25
Q

What is meant by the AI spring and summer?

A

In the late 1990s computing power and memory had increased. In addition, much more data became available. By cleverly training neural nets on these large amounts of data, AI suddenly became a reality. In the few years since that, almost all expectations have been exceeded

26
Q

What is meant by the singularity?

A

AI may become smarter than us. If so, there will be a point at which AI develops new AI itself. This would lead to a tremendous acceleration of AI development. This is known as the singularity.

27
Q

How does AI aid psychology? (3)

A

AI is increasingly used as a model for the brain (e.g. in the Brain and Cognition group).

AI is used to model large amounts of data (Data Science applications).

AI can be used to gather behavioral data previously unimaginable (e.g.COVID research Psychological Methods group)

28
Q

How does psychology aid AI?

A
  • Psychologists can leverage their research designs to understand AI better (e.g.subject AI to experimental designs)
  • Psychologists can inform novel AI-based techniques (e.g.emotion recognition, interpretation of results)
  • Psychologists can use AI in e.g.clinical and research applications as a predictive device
  • Psychologists can monitor the dangers of AI (bias, stereotyping, etc.)