Good day. Got up at 5:45, finished my morning routine by 6:40 then started work.

I feel less tired when I get up early, I think there's a sweet spot of sleep (around 8h for me) where if I sleep more or less I feel tired when I wake up.

Instead of counting the number of hours worked (~10h) I'm going to give a list of things I learned, since that's more useful to readers (and my future self)

• 2h Integrals, not much I can summarize as it's mostly procedural knowledge. I need to be more careful with my steps, I'm in the bad habit of checking my answers with Desmos when I should be differentiating them, then checking with Desmos as a last resort.
• 2h 30m Probability, realized inclusion-exclusion often has a symmetry making computation tractable, lots more procedural practice, watched two lectures radicalizing me as a Bayesian
• 1h Reading AIMA Bayes nets, Skimmed rest of section, will have to go back and read it thoroughly. Bayes nets are really cool though, and have tractable algorithms under reasonable assumptions
• 4h Reading the sequences and other posts, should have spent my time in a more focused way. Below is a summary of cool stuff I found
• ~1h Watching SoMe videos, specifically two on convolutions, the cerny conjecture, and negative probabilities

Stuff I read

• Timeless Physics - Sort of got the idea, very philosophically interesting, I want to read the rest of the Quantum Physics Sequence now, and learn physics in general. I've got to increase my focused-reading time and speed.

• The Dilemma: Science or Bayes? An example of Science conflicting with Bayes that teaches you about both

• Science Doesn't Trust Your Rationality Very interesting analogy between libertarianism and Science, and the pros/cons of both.

  Let us trust no one! Not even ourselves!

• Bayesians vs. Barbarians What would happen if a tribe of rationalists had to fight a tribe of barbarians? Would the rationalists be uncoordinated/not willing to do what it takes? The end sends chills down my spine

  So look at it this way: Jeffreyssai probably wouldn't give up against the Evil Barbarians if he were fighting alone. A whole army of beisutsukai masters ought to be a force that no one would mess with. That's the motivating vision. The question is how, exactly, that works.

• Writing Intelligent Characters Was good to skim, though purposeful reading is better. I'm not about to write any fiction, so I'd have been better off studying physics or something.

• That Alien Message Good thought experiment on AI and Solomonoff Induction, I should have been studying though...

• The Sin of Underconfidence An interesting comment:

  My usual self-testing example is something like "can I write this program correctly on the very first try?"

  I'm going to try this, reminds me of Don't write bugs

  I spent the next two years keeping a log of my bugs, both compile-time errors and run-time errors, and modified my coding to avoid the common ones.

  I need to try this in order to increase my programming skills. I can also give a probability to a program working then track results in a spreadsheet in order to calibrate my probability assignments.

  Using the computer to check code robs you of the mental exercise, perhaps watch-compilation and testing is harmful? I'm going to try writing decently sized programs without and see what happens.