Skip to main content

The Threefold nature of Intelligence

Image result for trefoil
Trefoil knot. Source: math.stackexchange.com
I am constantly contemplating Intelligence, not purely for the purpose of exercising my mind, but also to discover its secrets so as to enable myself or anyone I inspire to create an artifact that embodies it as it is inevitable that humanity will eventually build synthetic intelligence to augment its current capacities.


In my contemplations, I have summarized intelligence to be threefold in nature, with each aspect interacting inextricably with other parts in such a manner that it presents a singular view as a unified system.

We could go ahead and fracture this description to a great number of details as we try to assign nomenclature to every aspect of intelligent action that we can observe, but sometimes having a good generalized view is a powerful clutch to the thinking human as it distils the essence of the problem into a solid piece of abstraction which can enable the problem solver to make larger leaps in understanding without being hampered by excessive detail. I present this trinity of intelligence below with brief descriptions.

1. Memory: The first aspect of intelligence is memory. The intelligent system must be able to create an internal representation of the sense impressions reaching it from the outside in its most raw form as memory. Even though perfect memory is not a requirement of intelligence, the more raw information about the external world that an intelligent agent can internalize the easier and more accurate it will be to activate the other aspects of intelligence. This is the first aspect of intelligence.

2. Generalization: This is the second stage after the agent has internalized the external environment. It must now abstract away the noise in the input data so as to obtain information or the invariant features of this data. These invariant features of noisy input data are what we refer to as knowledge.

3. Generation: This is the output phase of intelligence where, from knowledge obtained in the generalization phase, the agent is able to generate data that either leads to external or internal actions. The central goal of the system will determine the kind of generalization and generation that a system will perform but this goal is not really a determining factor in the raw mechanisms of intelligence albeit it is important in directing the destiny of the system.

These are the 3 most important features of intelligence and although one can talk about things like planning, reasoning, intuition, creativity, etc. it can be somehow broken down into these three primitive functions.

For example, planning requires memory of the environment, invariant feature extraction like finding obstacles in some robot environment and Generation which determines the sets of actions that should be generated by the system in order to navigate the environment successfully which is the goal of the system.

In conclusion, any intelligent system will at least possess these three features although the details of the implementation will vary depending on the general goal of the system.

Comments

Popular posts from this blog

Next Steps Towards Strong Artificial Intelligence

If you follow current AI Research then it will be apparent to you that AI research, the deep learning type has stalled! This does not mean that new areas of application for existing techniques are not appearing but that the fundamentals have been solved and things have become pretty standardized.

Software of the future

From the appearance of the first programmable computer, the software has evolved in power and complexity. We went from manually toggling electronic switches to loading punch cards and eventually entering C code on the early PDP computers. Video screens came in and empowered the programmer very much, rather than waiting for printouts to debug our programs, we were directly manipulating code on the screen.


At the edge of a cliff - Quantum Computing

Source: https://ai.googleblog.com/2018/05/the-question-of-quantum-supremacy.html
Quantum computing reminds me of the early days of computer development (the 50s - 60s) where we had different companies come up with their different computer architectures. You had to learn how to use one architecture then move to another then another as you hoped to explore different systems that were better for one task than they were for another task.