How can we define Machine Leanning by itself?

Hi all:
I have a question.if the machine like TX1 can learn,why we have to train them?
if we trained the machine can we call it self learing?
so what the difference between learnning and classify?
how can I make TX1 learn all the objectives by itself without telling it 2 choose the correct answer from the trained data?

thanks

Suppose you are learning to recognize faces. A completely empty system has to be told what parts of images are faces. The more faces the system presents in which it has been told which parts are faces the more likely the system will get it right. It isn’t unlike a spam filter where it needs samples of what are spam and what are not before it can start seeing similarities in random cases without a human to help. Once enough samples are given in which someone tells the system the answers (the training) the system may become faster and better than a human at the same thing (think hundreds of thousands of images). It learns by being told what is right or wrong…it can’t know that without help.

What you are talking about is supervised learning with labeled training data. However there’s also unsupervised learning (autoencoders), semi-supervised learning, and reinforcement learning (DQN, DDPG, A3C) that learn from experience. You may be interested in these resources:

http://karpathy.github.io/2016/05/31/rl/
https://jkjung-avt.github.io/training-in-action/
https://github.com/vmayoral/basic_reinforcement_learning

thank you dusty_nv.