(WIP) Building my RL muscle
One of my goals in this year is to submit AI conference paper. (NOTE that it is not about getting accepted.) Personally, I am pretty interested in reinforcement learning (RL) even though it takes so much math background, which I do not have for now. While I am reading many RL papers to find research topics to write, I am writing RL codes at the same time. Of course there are tons of code out there I can simply copy and paste to do experiment about well-known environments such as MuJoCo....
Bayesian Statistics
Bayesian ๊ฐ๋ ์ด ๋์ค๋ฉด Bayesโ Rule์ ๋ํด์๋ง ์๊ณ ๊ทธ ๊ทผ๋ณธ์ ์ธ ๊ฐ๋ ์ ๋ํด ๊ณ์ ํท๊ฐ๋ฆฌ๊ฒ ๋์ ์ด๋ฒ ๊ธฐํ์ ํ๋ฒ ์ ๋ฆฌํด ๋ณด๊ธฐ๋ก ํ๋ค. ๋ฒ ์ด์ค์ฃผ์ (Bayesian) ๊ทธ๋ฆฌ๊ณ ๋น๋์ฃผ์ (Frequentist) Wikipedia์์ ์ ๋ฆฌํ Bayesian statistics์ ์ ์์ ๋ํด์ ์์๋ณด์. Bayesian statistics is a theory in the field of statistics based on the Bayesian interpretation of probability where probability expresses a degree of belief in an event. The degree of belief may be based on prior knowledge about the event, such as the results of previous experiments, or on personal beliefs about the event....
Hugo + github page ์ ์ ํ๊ธฐ
๊ฐ์ธ ๊ฐ๋ฐ ๋ธ๋ก๊ทธ๋ฅผ ์ธํ ํด์ผ์ง ํด์ผ์ง ํ๋ฉด์ ๋๋ฌด ๋ค๋ก๋ง ๋ฏธ๋ค๋๋ค๊ฐ ๊ต์ฅํ ๋ฉ์ ธ๋ณด์ด๋ ๋ธ๋ก๊ทธ(Lilian Wang ๋ธ๋ก๊ทธ)๋ฅผ ๋ฐ๊ฒฌํด์ ๋ฒค์น๋งํฌํ๊ธฐ๋ก ํ๋ค. Hugo๋ผ๋ ๊ฐ์ธ ๋ธ๋ก๊ทธ ํ๋ ์์ํฌ(?)์ github page ์กฐํฉ์ผ๋ก ๋ธ๋ก๊ทธ๋ฅผ ์์ํ๋ คํ ๋ ์ต์ํ์ ์คํ ์ผ๋ก ์ด๋ป๊ฒ ์์ํ๋์ง ์์๋ณด์. (์ฐธ๊ณ ๋ก Mac ํ๊ฒฝ์ ๊ธฐ์ค์ผ๋ก ์ค๋ช ํ๋ ๊ฒ์ด๋ค.) ์ค์น ๋ฐฉ๋ฒ 1. Hugo ์ค์นํ๊ธฐ brew install hugo 2. Hugo๋ฅผ ํตํด ์น์ฌ์ดํธ ๋ง๋ค๊ธฐ hugo new site qelloman-github-io ์์ ๊ฐ์ ๋ช ๋ น์ด๋ฅผ ์ด์ฉํ๋ฉด qelloman-github-io๋ผ๋ ๋๋ ํ ๋ฆฌ๊ฐ ์๊ธธ ๊ฒ์ด๊ณ , ์ฌ๊ธฐ์ ์น์ฌ์ดํธ์ ์ธํ ์ด๋ ์๋ก์ด ํฌ์คํธ๋ฅผ ๋ง๋ค ์ ์๋ค....