Learn (learning) the unlearnable
-in machine learning and eduction
How can we learn the unlearnable. The research topic of paper is
When must we learn the unlearnable?
how can we change the unlearnable to learnable.
I have developed the latter three questions for for some time, the paper entitles Learn (learning) the unlearnable, which will be submitted for publication.
How can divide unlearnability?. Available from: https://www.researchgate.net/post/How_can_divide_unlearnability#599379763d7f4bad493167b6 [accessed Aug 16, 2017].
Very interesting question and paper! I think to change the unlearnable to learnable is to begin designing it for human beings using learning theories. As educators, we apply the latter in teaching diverse learners.
Best regards,
Debra
Article Use of Facebook for enhancing student engagement in a higher...
Debra, Thank you for your comments:
1. changing the unlearnable to learnable is significant for educator in particular for primary education and middle education.
2. changing the unlearnable to learnable is also a topic for artificial intelligence and machine learning
3. changing the unlearnable to learnable is of theoretical significance for mathematics and wisdom development.
not sure if I get what you exactly mean but you may have a look at "stochastic models".
You may want to have a look at my publications. Those are based on Newell principles. We let the machine learning process generates some solvers that find some solutions for some specific problems.
At the start, I have attempted to automate a bit the human method of designing algorithm, starting from a random point.
Dear Mahdi
Thank you for your "stochastic models", which is not for unlearnability, the latter is related to a unlearnability space.
Dear Professor Patricia Ryser-Welch, to my knowledge, machine learning can only learn some unlearnable to the researchers, not to the whole human beings. This is the reason why I mentioned here the first question on unlearnability is classification of unlearnability based on the infinity of mathematics. Machine learning has been in the finite, learnable world, although it has been working for 70 years (since 1943). It has been in the infinite, unlearnable world, where I or we like to pursue. Deep learning cannot change this basic estimate.
Prof. Dr Zhaohao Sun
2017-8-18
Dear Prof. Sun,
that's an interesting question. Indeed, how can we learn the unlearnable. If it is truly unlearnable by definition, then I guess it cannot be learned. I assume, however, that many things being considered unlearnable can finally be learned. People might not be aware of their learning progress, yet their reactions still show that they have learned something about statistical regularities (such as in the associative learning / implicit learning literature). There are a number of machine learning approaches, e.g. artificial neural networks or Bayesian approaches that can approximate this type of learning quite well.
Please feel free to contact me should you have any further questions about this or should you need some literature.
Best wishes and kind regards,
Rainer
The questions of changing the unlearnable to learnable are discussed in the book
Prokopchuk Y.A. Sketch of the Formal Theory of Creativity. Monograph. - Dnepr : PSACEA Press, 2017. - 452 p.
(https://www.researchgate.net/publication/319331415_Sketch_of_the_formal_theory_of_creativity?fref=gc&dti=100257367027053)
Unlearnability is a matter of lack of creativity
Best regards,
Iurii
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning