My motivation is to learn and re-learn and use statistical physics. The entertaining pop articles are so much fun, but I also want to seriously explain something to someone who haven’t heard of it before (one test of understanding). Some questions I’m looking to answer: what is the model/technique about, assessing is it delivering what it promised, are these methods appropriate in my question. Below are my first explanatory pieces on “the ways of seeing the big thing from many small things” (statistical mechanics).
Below are my paraphrasing and questions and notes from NS162 class.
Before leaving Seoul, I had a wish to fill my regret of not fully digesting statistical mechanics from the 2AM + numbers-heavy class. My current perception of this field is that is deals with problems that have many many parts but can be described with equilibriums and averages (see intro to complexity). It was invented for making steam engines, later appropriated by mathematicians, physicists and computer scientists to find or solve problems of their interests (for example, Leo and Howe were Q&A’ing about this, maximum likelihood in Bayesian statistics is translatable with energy minimization in statistic mechanics, just in probability lingo instead of physics lingo).
First I will summarize some frameworks and notes from that class, then I will flesh out the parts that are more relevant because I want to discuss with Yut and Ellie for their contexts, and I want to figure out how I’d relate the data science X physics project back to learning about the newest coolest best methods for humans understanding and doing the world (this is how I mythologize the umbrella of complex systems or contemporary science styles).
What is the best approach to learning stat mech?
First off, framework of the topic can be summarized by either the list of LOs or textbook table of content.