Lecture3-1Learning with different data label

Supervised Learning

Unsupervised Learning

Semi-supervised Learning

Reinforcement Learning

A very different but natural way of learning

Teach your dog: SIT DOWN

cannot easily show the dog that yn= sit down, when xn= ‘sit down’

• The dog pees on the ground. (BAD)

  • punish to say yn~=pee is wrong.

• The dog sits down. (GOOD)

  • reward to say yn~=sit is good

Mini Summary