- [[dimensional reduction]]
- [[Dimensionality reduction is an abstraction]]
- There is no objective reason for choosing which axis to crush. - Abstraction is subjective
- Then it’s not as if you randomly select an axis.
- If there is something to be gained as a result of choosing the appropriate axis of abstraction, then choosing that axis is reinforced.
- That is, reinforcement learning.
- Which axis to choose multi-armed bandit.
This page is auto-translated from /nishio/強化学習と次元削減 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.