Can Artificial Intelligence have Creativity.
It all depends how we define Creativity.
Creativity is a mental and social process involving the generation of new ideas or concepts, or new associations of the creative mind between existing ideas or concepts. An alternative conception of creativness is that it is simply the act of making something new.
Surely we can just create a machine and have it output random numbers but that would be more like a retard's creativity than an artist's creativity.
In order to complete the process we also need analytical thought which would in-turn judge the application/use of the thought.
As seen in the video, our brains(right side) randomly generate thoughts due to noise generated by heat.
And then these thoughts are processed by our logical analytic part of brain(left side)
to decide if it is of any use to process it.
Clearly the logical analytic for humans can be linked to quering from a huge database of knowledge.
New borns do not have that and hence rely on adults to do teach them. Same is the process with AI.
Another key requirement lacking currently is the ability to gain knowledge, learn knowledge and curiosity(interest) to learn.
Again the field of curiosity is again a result of creativity and analysis deciding which field is more beneficial for them on a sub-conscious level.
So, what I believe is that we need to research in generating algorithms for analysis.
I am afraid the organisation "In Its Image" does not seem to have attained it as they claim to have it since 1970's.
It is a complete scam since such a breakthrough in technology is in demand and still researchs are going heavily by huge companies like IBM
and they say that all they have attained so far is the intellegence of a drunken cockroach who takes an 30 minutes to figure out his way out of a room.
But it sure gave me ideas.
Being a software programmer, I do try to write programs that have basic intellegence.
For eg: An operation is being carried out. Error occours. The program retries. Error occours again. If its the same error, it gives it a final try and if it happens again, its informs the tech support by emailing/paging them.
If another error happens, it gives it another try. If different errors happens it tries until 10 errors happened in succession before contacting support.
(Don't worry, its not a critical process that needs attention for every error).
What would be AI in my understanding, it keeps an error log. And refers that everytime an error happens.
If the error log suggests that everytime Error 'X' happened, it kept happening again and again, It would stop and contact support.
Even better would be
If there are any situations that Error 'X' happened and did not happen in second try,
there would have a random bit deciding if it wants to give it a try or not.
The chance of randomness would be affected by number of time Error 'X' happened before and corrected AND number of time Error 'X' happened again.
This process gives the program ability to grow its database.
Since the chance of selecting an error happened earlier depends on previous behaviour, the program keeps getting wiser.
It sounds complex, but is no comparison to the calculations that take place in our minds deciding if we want to take Road 'X' to work or Road 'Y'.
But they are on similar grounds.
Still, the technology still has hurdles with space of storing database to speed of processing queries.
And what lacks most is the reliability of human judgement which is achieved by huge database of our experience/knowledge and complex queries we do for our decisions instead of just few like this program.
But creativity is complete different game.
Our program can only decide in case of an event. It does not and figure out way to fix the error.
Hence, all we acheived was analysis, not creativity.
In order to have creativity, we would again need a pre-defined database for quering the usability of the thought.
In the case of our program, it would be like using the existing database, and while the program is running without errors, the program in the background,
searches for most common errors that occour(field of curiosity), keeps making patterns of errors happening in succession(creativity) and then refffering the source code to find out possible causes of errors(It would query another database with list of errors and possible causes and solutions). With a list of possible causes, it would go ahead and implement the changes to a testing program checking to see if it fixes the errors from happening again(analysis) and do some testing, make sure its run a couple of times. And then edit the real source code and save it.
Thats when I think the complete process of creativity would be complete.
But could we device a system that could query the causes of erros and their possible causes & solutions and create solutions for unknown errors?
Means combine, defragment existing solutions and create new better solutions by testing their efficiency.
A constant need to improve.
Just like this post, every bit was written as it came to my mind.
Let me know what are your thoughts
I was eager to write my own thoughts after watching the documentry: In Its Image
Be the first to rate this post
- Currently 0/5 Stars.
- 1
- 2
- 3
- 4
- 5
Tags:
artificial intelligencem,
ai,
creativity,
creative machine
Categories:
Technology