Tag Info

Hot answers tagged

8

A (probably incomplete) list of ongoing whole brain simulation projects can be found at http://www.artificialbrains.com/. However, based on the information reported on this site, it is sometimes hard do distinguish what already has been achieved and what is still in the planning stage. Nevertheless it gives a good overview to start with.


8

Actually, standard IQ tests, such as Raven's matrices, tend to assess intelligence better if they are not timed. In this paper by Philip Vernon (1988) it was found that the g-factor extracted slightly more variance for the same test if the test was not timed than if it had a time limit. This means if you ask yourself: "What is this test measuring?", you can ...


7

There are two ways to approach your question: with or without dualism. I will highlight the dualist approach since it is more salient. Keep in mind that I do not find this approach reasonable, and doubt my summary will do it justice. You might be interested in the concept of philosophical zombie, it is a modal argument against physicalism in the spirit of ...


6

As per the comments to the question, human research observing this distinction does exist. CHCH possibly alludes to an article by Gläscher, Daw, Dayan and O'Doherty (2010) which concisely defines the difference between model-free learning and model-based learning: Reinforcement learning (RL) uses sequential experience with situations (“states”) and ...


5

You are certainly not the first to conceive of intelligence as pattern analysis. In fact, there is a book by Jeff Hawkins called "On Intelligence" which investigates this idea in depth. His idea is that the best candidate mechanism for intelligent brain function is predictive coding. In the predictive coding framework, the job of the brain is not to ...


4

A human brain recognises letters by their constituent features (sub-letter parts). It is modelized by a pandemonium model where printed information is extracted locally then globally. In the letter recognition literature, this type of feature-based hierarchical model competes with template matching theories (with an advantage to the pandemonium-like models ...


4

An AGI is a man-made machine that can learn, adapt, think, plan, predict, etc. Cognitive science is the study of how our "biological machines" do those same processes.


4

Cognitive Architectures The description most closely matches the concept of a cognitive architecture. Whereas I would say most empirical cognitive science focuses on isolating cognitive functions or behavioral substrates, cognitive architectures are relatively unique because they attempt to run bottom-up simulations of interdependent sets of cognitive ...


3

Cyc is similar to what you are looking for. It is a database of general knowledge and is organized semantically. It is free to use for research purposes.


3

Interesting question! A related phenomenon called the illusion of explanatory depth (IOED) suggests that the human cognitive system has a systematic weakness in this kind of evaluation--I believe the classic example is asking people if they know how a helicopter works (most people say yes), and then asking them to explain how a helicopter works (very few ...


3

Depending on the definition of "psychology" and "psychological problems" sure. The APA defines psychology as "the scientific study of the behavior of individuals and their mental processes." If we take this to only apply to biological systems (as I am sure many researchers do) then no, because the robot is not a biological system and therefore we can not ...


2

I think "teaching of high-level strategies will allow students to use learned strategies across different domains" is the very rationale of mathematics. Math gives very good examples of both abstract strategies for solving problems across different domains and also specific, explicit strategies. You don't seem to want the abstract strategies, but if you ...


2

I'm not sure I fully understand your design; perhaps you can clarify what you want your network to learn, why TD-learning "isn't cutting it", and what you mean by 'reinforcement' and 'prediction' learning. In particular, TD-learning is a reinforcement learning model, and it does reward based on predicted (and not just observed) outcomes. However, you seem ...


2

Short answer: We don't know. Long answer: There are a few major lines of thinking on the subject currently. Cognitive closure: One common argument is that this question is simply not answerable - at least not by humans. By this view, it is possible that the creation of an artificial intelligence that even resembles humans sufficiently to suggest ...


1

Empty the small jug. Fill the large jug. Use the large jug to fill the small jug. This leaves 2L in the large jug and 3L in the small jug. Empty the small jug. Use the large jug to fill the small jug. This leaves 0L in the large jug and 2L in the small jug. Fill the large jug. Use the large jug to fill the small jug. This leaves 4L in the large jug and 3L ...


1

The distinction isn't rigid, but it has to do with whether the program controlling the robot is designed to mimic the human mind in some way. If the robot is a model of human cognition, then the term cognitive robotics is sometimes used. Artificial Intelligence is a general term for software that behaves in an intelligent manner. There are plenty of ...


1

I feel that the label of consciousness is merely a semantic distinction that belongs to the realm of philosophy, not neuropsychology. Like Noam Chomsky mentioned in one of his talks hosted by Lawrence Krauss - we could also ask ourselves whether animals (e.g. dogs) are conscious. I'm not exactly sure, he mentioned that birds are said to "fly" in Enlgish but ...


1

This question has perplexed me for quite a while now. The problem with declaring an artificially intelligent machine 'conscious' is the very definition of consciousness. A quick google search for the definition for 'consciousness' returns 'the state of being awake and aware of one's surroundings'. This definition in my opinion is too vague to be extendable ...


1

So I guess the question is whether such a test can be constructed at all. In other words, is there a category of problems that require some minimal intelligence level to solve them no matter how long one might try. Like the problem that could be solved in a minute by a smarter person but is impenetrable to you no matter how many years you spend ...


1

The nature of intelligence is a highly controversial open question. However your phrasing and use of the term "general intelligence" seems to indicate a g-factorist context to the question, so then, I would say the answer is "no". Pattern recognition tests such as Raven's Progressive Matrices do load highly on the g-factor but things unrelated to pattern ...



Only top voted, non community-wiki answers of a minimum length are eligible