3 ms·
I think this is a huge part of it. Realistic hacky workaround is to get one of those "1000 calculus problems solved" books, and work through the examples -- th
by shanusmagnus 4y ago
I think this is a huge part of it. Realistic hacky workaround is to get one of those "1000 calculus problems solved" books, and work through the examples -- there's a zillion books for math topic X online, you can read through the relevant parts of a few. But your progress / insight will be in proportion to the problems you actively engage with.
I've often thought that a really pedestrian, but possibly world-shaking, application of AI would be generating math problems (and solutions) appropriate to someone's state of knowledge. Such a thing would make problem-solving practically as fun as a game, at least for many people. Imagine the consequences of a generation of kids obsessive solving of interesting multi-area math / engineering problems the way recent kids played Minecraft or whatever.
(Q: Is there anything vaguely like this in the world now?)
- dontbenebby 4y agoIf you know a good resource on learning calculus on your own, I'd add it to my todo list. Every algorithm I reverse engineer or develop has been very... I think linear is the phrasing? No curves? >I've often thought that a really pedestrian, but possibly world-shaking, application of AI would be generating math problems (and solutions) Maybe read up on plagarism detectors? I can't recall the title but there was one some people used in grad school to spot code re-use that was engineered much differently from Turnitin. (Play around with the latter, and notice how the "plagarism" score jumps when you tell it to ignore what you cited. I had way too many stories told to me of professors who'd start the academic misconduct proces because they forgot to do that, paired with witnessing some egregious academic misconduct that I'll simply never be able to get away with.) - Greg.