@Ektelion14@lorel_sg@lmrlbastos On peut donner comme exemples pour les cas où Newton ne marche pas les classiques périhélie de Mercure, effets de lentillage, redshift gravitationnel, dilatation du temps, etc... même si comme le mentionnait Lorel, ce n'est pas très utile face à du slop IA.
@Ektelion14@lorel_sg@lmrlbastos Il n'a pas qualifié la théorie de Newton d'invalide. Ce qu'il a dit, c'est qu'aujourd'hui elle est "invalidée dans plein de cas". Ce qui est tout à fait juste (il est d'ailleurs bien placé pour le savoir étant données ses recherches sur des théories modifiées de gravitation).
As a sophomore, I took Probability Theory with Yasha Sinai.
At some point, he was explaining something, looked back at us, and realized we were all lost. He said: “If you are confused, it is not your fault; it is mine, for not explaining it clearly enough. Let’s try again.”
I was flabbergasted. Here was one of the greatest minds of all time, apologizing to a bunch of 19 year old idiots that they can’t follow his argument. Wow! I always think about that, and try to live up to it: if my students don’t understand something, then it’s *my* fault, not theirs.
Which brings me to Lecture 12 (posted here: https://t.co/kUaTLT1U1s). I tried to do too much at once. We were proving that if a sequence is monotone and bounded, then it’s Cauchy. (En route to Bolzano-Weierstass, and the — hopefully well-motived — construction of the real numbers!) In natural language, there is *so* much handwaving with these types of arguments, “just make n and m sufficiently large; clearly you can do that”. Yes, it’s “clear”, but … why *exactly*? Lean doesn’t understand handwaving.
The key idea is to use orbits. But I made the mistake of introducing orbits *in the middle* of an otherwise already complicated proof. I didn’t realize that orbits themselves would be a new concept to them. Pedagogical mistake, not the right scaffolding. So I’ve refactored the game (too late for this lecture, but hopefully will help the future) and lecture notes:
https://t.co/dzu9FE0Jeb
Live and learn!
@douzedouilles En supposant un taux de mélanine élevé et les molécules composant le cuir du cheval toutes dans leur état fondamental à t = 0, on peut calculer l'énergie moyenne de transition entre le ground state et les différents niveaux excités. On déduit ν = E/h la fréquence -> il est blanc
@0xRy4n@AyoolaMatthee@InternetH0F Good point. I really dislike this way of making the customer implicitly obligated to pay more than what they calculated just because the system doesn't pay decently its workers, but it still is a custom in a country I am only visiting, I'll abide to it out of respect