Sunday, November 26, 2017

Artificial Intelligence --- Machine Learning and Deep Learning

Maybe an alternate subject line:  Looking at the AI forest and seeing something deeper in the woods and they arn't the trees.

The thought popped into my head this morning.  I woke up too early.  Now I have to go to Google to flesh it out:  https://www.google.com/search?client=safari&rls=en&q=artificial+intelligence+machine+laerning+deep+learning&ie=UTF-8&oe=UTF-8

Machine Learning:  (ML) https://en.wikipedia.org/wiki/Machine_learning

Deep Learning:  (DL) https://en.wikipedia.org/wiki/Deep_learning

Deep Learning goes deeper!  Duh!

Deeper or Higher?  Both ways maybe?  Framing it is everything?

Deeper: My own definition: The process of going from higher to lower level.

Higher:  (Self evident definition!)

Higher Learning?  That is done at universities.
http://www.complex.com/pop-culture/2015/01/higher-learning-20th-anniversary
In John Singleton’s eyes, to "unlearn" means to release the toxins which pollute individuals and society. But that cleansing process also involves adopting a new, more progressive set of values. Which leads to the bigger question: What have we learned in the two decades since Higher Learning premiered?

Perhaps the value of AI Deep Learning is to teach us what to Unlearn?  Unlearn the toxin of what we feel and restructuring that empty bowl with what we logically think and know.

AI goes deeper....Human Intelligence (HI) and Human Resources (HR) go higher?

I think that AI has gone down to the most granular elemental level of conceptual things to work with maybe 80% of everything in our physical environment being modeled on a machine and maybe 60% of our human conceptual environment.  AI can go deeper but at the same time it has enough basic elemental things to start to put them together and go higher in learning as the same time it goes lower.

Going two ways at once is a hard thing to grasp.  Maybe a better concept to describe the current state of ML and DL is Yin and Yang:https://en.wikipedia.org/wiki/Yin_and_yang. 
Not binary but relational.

There is a fear of machines taking over human jobs.  A fear that they will be smarter than us.  Perhaps that is like being afraid of a teacher that is smarter than us?  Rejecting the teacher.  Not wanting to learn?

Logic and Reason is the best teacher. We have much to learn from a AI Deep Teacher that by our own design is only one step ahead of the student.

Wadda ya know (?) (!)

What do we have to fear in learning?  How ignorant we are?  That is what Higher Learning cures.

What we learn empowers what we do.  In the Information age what we have is a logical progression of "Book Learning"? http://www.thesaurus.com/browse/book-learning which is really learning theory for us to put into practice to serve ourselves.

Book Learning?  Think about it.  Highly ambiguous in meaning we do not learn from a Book any more than we learn from a Machine.  They are merely the structured Repository Medium what is to be learned delivering a Message to us with a payload of giving us new Attributes, Methods to apply them and Messages to implement them.

Machine learning is Book Teaching 2.0

Deep Learning is Machine Teaching 2.0

We are always the students of our own devices that enable us to learn more.

Hello Watson! 

https://www.technologyreview.com/s/513696/deep-learning/

https://www.technologyreview.com/video/609388/ais-language-problem/

No comments: