Principles and Applications of Reinforcement Learning

View
Some learning problems don't have "yes" or "no" answers or even clear-cut numerical answers. Many of these are optimization or strategy problems where each choice works better than others in some ways and worse than others in some ways. In reinforcement learning, agents can choose actions to gain rewards or punishment for choices, thereby learning what works and what doesn't in problem-solving.

Source: Martin Hilbert, https://www.youtube.com/watch?v=2sg_2_ltoUU
Creative Commons License This work is licensed under a Creative Commons Attribution 3.0 License.