This is a vague question. I will do my best, I think it has definite answers. I am hoping for answers of the form "Read book x, learn this specific topic, read this paper/s".
What is bothering me is that every AI system I have read about seems to have an internal model which is set up to operate in a specific way. The task of the learning algorithm is to then optimise the parameters of this model to solve the problem.
I am thinking that this is akin to setting up a custom description language which preserves or translates some structure of the thing to be modelled in such a way that it becomes amenable to solution through some understood algorithmic technique; iterative gradient descent, analytic, whatever.
The Question: Is there any framework in which the optimum model/description language can be discovered? For instance in a GA the mapping from the genome to the phenotype is critical to the effective operation of the algorithm. This mapping is manually defined and, I would posit, is the most critical component of the GA. Specifically how would one search through the space of mappings? Or going meta how would one search through the space of algorithms for finding that best mapping? It feels like you could dissapear up that ladder forever. In order to solve the problem you must first solve the problem.
Neural networks seem to represent an effective model for solving real world problems, why? How would one generate another model for a different set of problems? Cellular Automata seem good for modelling traffic flows or disease transmission, why? How would I automatically generate alternative systems which also model the set of problems effectively? etc.
Is this something that is thought about and discussed? Where would I find it?
Hi,
That is a very good point. If you read about the no Free lunch theorem, you will see that a method may only work for one specific types of problems.
I have attempted to define a generative framework based on three layers. One for the problem, one that learn an algorithm that solves the problem, and a third one that passes the information about the problem domain and the learning process. You will fine the formal definition in my published thesis.
This aspect has received some mixed feedback from the genetic programming and also the hyper-heuristics communities. They appear not to be interested in some type of languages.
The optimisation of parameters has been studied for many years and some frameworks have been developed 15 years ago. Neural network is very good at mapping some patterns to some certain outcomes. However, it would be interesting to investigate how the genetic programming and the neural network communities could work better together.
I would also look at the automated software engineering. This is again another research area that may share some elements with generative neural networks and the genetic programming world.
In my opinion, we need to talk more to each others.
P.
I do not see how this can be done for more than a VERY restricted type of problem setting, even with very limited ranges of data. The world of problem instances is - for better or worse - simply too vast to be able to map into something simple.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence