5
$\begingroup$

Newbie here.

I recently read about cognitive architectures (see: https://en.wikipedia.org/wiki/Cognitive_architecture). They are supposed to be modeled after the human mind and represent a promising approach towards artificial general intelligence (AGI).

My question is, however, why haven't these cognitive architectures achieved AGI yet? What are the specific limitations and roadblocks that cognitive architectures face?

Improve this question
$\endgroup$
5
  • $\begingroup$ IMO the problem is we don't know how to train giant recurrent neural networks over very long timescales. There is likely some giant recurrent neural network that could do the task of AGI, running on current hardware - but we don't know how to find the weights for it. Gradient descent suffers from the vanishing/exploding gradient problem. $\endgroup$
    – causative
    Commented May 12, 2021 at 4:00
  • 1
    $\begingroup$ Another way to look at it is, if you manually design a large software system such as a cognitive architecture, it's going to be full of bugs, because all large manually designed software systems are full of bugs, especially when they must interact with the real world. Thus what we need is some way of automatically ironing out the bugs. This is what gradient descent can do for neural networks, but it just doesn't do it well enough, and many cognitive architectures can't be backpropagated end-to-end anyway. $\endgroup$
    – causative
    Commented May 12, 2021 at 4:06
  • 1
    $\begingroup$ @causative: a possibility could be to generate the code of that cognitive architecture software. This was suggested by Jacques Pitrat in his Artificial Beings: the Conscience of a Conscious Machine book, and with others I am trying to reimplement that idea for the RefPerSys project. However, such an approach needs years of work. Feel free to email me at [email protected] $\endgroup$
    – Basile Starynkevitch
    Commented May 12, 2021 at 5:54
  • $\begingroup$ @Anonymous: I tend to believe that a full book could be written to answer your question. Did you consider starting your PhD on it? $\endgroup$
    – Basile Starynkevitch
    Commented May 12, 2021 at 5:57
  • 1
    $\begingroup$ The key problem is understanding how humans and other creatures achieve a level of general intelligence. IMHO, unless we just stumble across something out of luck, until we understand and can define that in all its detail, we will be unable to define what is missing from our attempts so far. $\endgroup$
    – David Hoelzer
    Commented Apr 18, 2023 at 13:19

2 Answers 2

Reset to default
1
$\begingroup$

This is a philosophical question that does not have a one-off answer.

If I may suggest a quick thought:

Our brains are trained on millions of tasks when we grow up (recognising so many objects, emotions, actions, etc.), that if we were to make one giant neural network, and we train them on so many tasks, perhaps we can achieve a human like response.

As for 'consciousness'. Do any of us really know how that works? Is that due to the neural networks in our brain or something more intangible (that spark of life)? I'm not so sure of it, but it surely seems more complex than any of the deep learning models I've seen.

Improve this answer
$\endgroup$
1
  • $\begingroup$ This is not necessarily a philosophical question, if one precisely defines AGI and how to measure it. So, it could be a scientific question, but we probably don't know the scientific answer to the question. $\endgroup$
    – nbro
    Commented Apr 18, 2023 at 21:54
0
$\begingroup$

That article only describes a way of thinking about and modeling the sort of cognitive capabilities that can lead to AGI. The thoughts and models are in no way a clear and certain roadmap to achieving AGI. And, even if they were, there are many implementation hurdles that would still need to be cleared to realize this.

Grady Booch summed this up pretty well in a recent tweet:

https://twitter.com/Grady_Booch/status/1615284029594697728 We do not yet - nor do I expect we will anytime in the future - have the proper architecture for the semantics of causality, abductive reasoning, common sense reasoning, theory of mind and of self, or subjective experience.

Improve this answer
$\endgroup$

You must log in to answer this question.

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