7
$\begingroup$

Given the following definition of an intelligent agent (taken from a Wikipedia article)

If an agent acts so as to maximize the expected value of a performance measure based on past experience and knowledge then it is intelligent

and given that we, humans, all make mistakes, which means that we are not maximizing the expected value of a performance measure, then does this imply that humans are not intelligent?

Improve this question
$\endgroup$
0

2 Answers 2

Reset to default
2
$\begingroup$

It rather depends on how one defines several of the terms used. For example:

  • Whether the term "expected" is interpreted in a formal (i.e. statistical) sense.
  • Whether it's assumed that humans have any kind of utilitarian "performance measure".

The motivation for this description of "agent" arose from a desire to have a quantitative model - it's not clear that such a model is a good fit for human cognition.

However, there are alternative definitions of agents, for example the BDI model, which are rather more open-ended and hence more obviously applicable to humans.

Improve this answer
$\endgroup$
4
$\begingroup$

"the human mind is a battleground of higher level goals and lower level goals "
— Marvin Minsky paraphrasing Sigmund Freud

I argue that in general human agents try to maximise a hierarchy of performance measures.

performance measures of humans

  • Survival of genetic data

    • Energy supply and Water

    • Sex

      • myriad subgoals....

Mysterious mental mechanisms which neuroscientists do not understand yet force the average human agent to maximise various evaluation metrics. With the overarching goal of survival of genetic information. Successful genes are immortal. We are still under the yoke of an ancient genetic algorithm.

These measures are optimised throughout a humans life time. A 30 year old agent is better at survival than a 10 year old agent. A 30 year old agent makes fewer mistakes.

We remember our mistakes. Mistakes are burned into our memory by high levels of neurotransmitters (and reinforcing of synapses) so we don't make them again.

We attempt to optimise a swarm of subgoals that are all connected in one way or another to the main goal gene survival.

  • status

  • money

  • education

  • happiness

Improve this answer
$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .