(The below text version of the notes is for search purposes and convenience. See the PDF version for proper formatting such as bold, italics, etc., and graphics where applicable. Copyright: 2022 Retraice, Inc.)
Re64: Math, Code and Deadlines
(AIMA4e planning)
retraice.com
Coming to terms with the hard things first.
Air date: Monday, 28th Nov. 2022, 11:00 PM Eastern/US.
How long do we have? Deadlines
We're going to do a deep-dive on Russell & Norvig (2020). Mastery should be of a discipline, not a particular book, and will require returning again and again to AIMA4e as real-world tests reveal more and more of its utility for that purpose. How long to deep-dive?
* A year? No, six months, with prereq month, December, for the hard stuff (math, code);
* Reality isn't going to wait; recall the deadline problem;^1 and we should probably never make a plan longer than six months.
To do what? Calibration by tests
Tests can be self-imposed or world-imposed, and toward calibration (Do I know what I think I know?) or toward gatekeeping (school, graduation, etc.).
Our self-imposed tests...
* competence as tested by the online exercises, and comprehension of the technical details in the text itself;
* a survey of the whole field of AI is the larger objective, as tested by published paper comprehension.
Later, world-imposed tests...
* real-world use tests will be toward mastery;
* mastery of a subfield (specialization) will be tested by action in the world--`play', `games', wins and losses.
And Retraice? Learning vs. teaching
I'll be learning, not teaching. "What's going on out there" is computer control. I'll be like a remote correspondent(?), or beat reporter(?) on that current history. But we shouldn't decide the best form Retraice can take, we should be open-minded and discover it.
What strategy? Hard first: Get out of English quickly
We're not expecting difficulty with the English in the book. It's the code and the math.
Code: ABC: always be coding. Math: resources at the ready (the best books).
* Code: TABLE-DRIVEN-AGENT, p. 48;
* Math: fate of the table agent, p. 48;
* Environments (Linux, Python, LaTeX, Github, etc.) for:
+ study (see, and report back);
+ output (do, and report back).
__
References
Retraice (2022/10/16). Re20: The Deadline Problem. retraice.com.
https://www.retraice.com/segments/re20 Retrieved 17th Oct. 2022.
Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Approach. Pearson, 4th ed. ISBN: 978-0134610993. Searches:
https://www.amazon.com/s?k=978-0134610993
https://www.google.com/search?q=isbn+978-0134610993
https://lccn.loc.gov/2019047498
Footnotes
^1 Retraice (2022/10/16).
