Видео

Having a class in school where you just play puzzle games for an hour would have been great

4 with blanks, 2 without blanks

16:20 Hold up... Imagine the feel of what now? 💀

GUESS: Givens, Unknowns, Equations, Substitute, Solve ❌ GUESGS: Givens, Unknowns, Equations, Substitute, Guess, Solve ✅ Guess Inception

I agree with most of the conclusions made in the video. But when I think of cooking in terms of learning science, cooking becomes an activity where I must achieve more and more competency in, and it almost becomes an objective rather than a necessity, which I see as redundant. From my perspective, I don’t understand why you would try to improve your cooking skills beyond basic competency to cook for your family. I can understand trying to optimize your skill set for work, but not for housework like cooking. My overall criticism is that it just seems redundant to apply learning science to cooking when u can already cook just fine & feed urself.

Very good. We need more in different subjects like physics or statistics.

Oh, hey, I do this! This is a generally applicable approach to building intuition/understanding. All you have to do is *always* make a prediction before you gather concrete information. For a simple example from my life, I always guess what time it is before I look at a clock, and because of that my internal clock is pretty damn sharp.

Great guidance but how can I know if what you just taught us is limiting me from looking at problem solving from another perspective

Taiwanese viewer here 👋

I learned how to cook quite early in my life: my approach was putting everything together what was unheard of def helped me understand concepts behind dishes rather than boringly praying down recipes (I really despise that)

I've studied the tower of hanoi in algorithms

What are the traits of highly intelligent people? Can we train ourselves to be one of them? It would be so interesting if you take these questions as an idea for next video!!!

I want you to know that I plan to take study techniques courses to improve my grades and my time studying, learning affectively in the long term, I have reviewed several videos and platforms and found 2 that convinced me because of their quality, these are cajun koi academy that has a course called study quest and justing sung with his icantudy course, the latter is expensive in relation to dollars and is in English, but offers discount scholarships to people with poor economic situations. I tell you that recently I am going to have a new little brother and with this the economic situation of my parents will be affected, I already showed them the courses that I want to take but my parents did not say yes or no but for the moment it is under discussion, besides the fact that we do not have a credit card and my country has a shortage of dollars and I do not know if my parents will accept. I would like you to tell me your opinion on this matter with my current situation, and give me some advice since I am about to finish my first semester in medicine and I will have a month off, but the second semester will be more difficult and that is why I wanted to take these courses.

this may sound unrelated but can you please make a video about what reading books actually does to brain that makes reading so important. I also wonder how reading improves problem solving skills, critical thinking etc.

This is so important

Is this only applicable to learning physics or also applicable to subjects like law?

Any guides as to figuring the expert skills in any domain? Thanks!

Wow. Solving thousands of problems in physics without learning physics sounds a lot like the real life analog to overfitting, a concept in machine learning. In machine learning it is common practice to separate the data out into multiple sets. A training, validation and test set. They're trained with the training set, of course, and during that time, their progress is assessed with the validation set. What this does is it prevents the model from essentially memorising the data points which wouldn't be really useful in the real world. But the validation set also isn't perfect because it is used to inform the ML practitioner's decisions on how to train things, and eventually they might accidentally land on a model that miraculously understands the underlying pattern... but only for the training and validation sets. So you have to test that as well. Chances are because of the numerous iterations and tweaks from the practitioner, the models that seem the best are likely to have a positive bonus purely from chance. They go through that much for an extra few percentages of accuracy. Compare this to the education system which holds practice tests as bibles. "Guy's it's simple, when we give them practice tests, they do better!" There were literal companies coming in to give a basic talks on ways to increase your grades at my school. And it boiled down to taking practice tests. That was it... Teaching to the test. Perhaps that's why the education system is so wanting - because they inform themselves based on the tests and only the tests. No question of true performance or understanding. This doesn't seem changeable because on which metrics do we evaluate a better education? It's like a misaligned AGI, the government. Can't tell it nothing except what will get it closer to it's predefined immutable goal. I know it may be somewhat a stretch to say that they're teaching to the test, since the practice tests are actually different data-points, so to speak, to the real tests - but call them training papers and it's the same thing. The environment is the same, so you're not testing conceptual understanding necessarily, so much so as you are testing paper doing. And certainly papers have their patterns.

I will be sharing this with my A-level Physics students (In China) next semester Many thanks!

Great video! Appreciate your work 🙌🏼

Instructions unclear, I now have liver cancer because I drink alcohol after every study session.

Can an example be given? Like for something like linear algebra, or programming/algorithms? I am not sure how to apply that step..

I wonder if he likes Go...

This is a great use of logic for problem solving. But this does not limit using this concept to hard science. I teach people how to drive a truck, to get a Commercial Driver License. The toughest thing to learn is how to back a semi-truck properly. I tell my students the first thing to do is to imagine where the back wheels will be when they are done making an adjustment in their backing problem. Then how to get them there. For me, that is the point of Dr Keep's video - imagine the solution before solving the problem. (I was completely distracted by the Go (wei-chi) board on the stand!)

It sounds like this procedure is widely applicable even to abstract problems such as marketing or art. Trying to predict before executing, e.g. running an A/B test for marketing or painting for art. It sounds like you'd be creating a habit of gauging first whether something is good abstractly - using those abstract ideas to guide your decision making throughout execution phase. I suppose for marketing you'd focus on ideas like "how much of and which of these demographics are affected and how so?", "what might this person say about this brand for this particular webpage?", or "would this audience stay longer or shorter on the site when they first see this?" and art would be about visualising it, asking questions about different parts like "would this make the face look happier?", "would this make the painting more vibrant?", or "would this contrast poorly with that?" I wouldn't know much about some kind of "learning loop" you'd employ. Mr. Keep, I wonder about your thoughts on working through abstract problems that don't have concrete answers as do the perfect worlds of maths and physics. What do you think about the application of mental models in step 2.5 of prediction? Things like inversion? "What could I do to make the potential customer not want to buy at all?" "What simple small thing could I do to ruin this painting?" Or consideration of the extremes? "If X goes to say a billion what does this look like?" And so on...?

This video is legendary thank you

phenomenal content, thank you!

You are like the absolute gold standard channel on learning and its sad you're criminally underated!! Keep making these pls!!! (Anyone reading like the video)

Very rare quality content video

This is the genuinely most fascinating channel I have ever come across! Thank you for sharing your gift. If anybody hasn't signed up for the newsletter, it is similarly a goldmine!

Thanks for the video, really helpful stuff, I hope I'll remember to come back to this when the semester starts haha. On a side note, I wanted to ask your perspective on 2 questions about practical implementation of learning science into real life. I'm sure you're busy, so you can answer with as little or as much detail as you like (or maybe make a video if that seems like a good idea!). 1) Do you have a structure or recommendation for going about implementing all the ideas and advice in your videos into our personal, real life pursuits? Some sort of way to identify and adapt relevant theory to whatever thing we want to learn, whether it's physics, social skills, starcraft, stress management, or even something as simple as building an exercise habit? I guess you could call it meta-meta-learning haha. Context: I find myself watching a lot of your videos and feeling very enlightened and captivated, but looking back I haven't actually put as much of it into practice as I'd like. I think it's because it can be difficult to know where to start, which concepts to apply where, and things like this. 2) What's your opinion on how emotional intelligence plays into learning, and how can you practically improve it to make the actual (and hopefully more effective) practice you end up doing in real life adaptable enough to your preferences and the follies of everyday life to be enjoyable and somewhat sustainable? Context: There's a lot of emotional traps that you can fall into while learning that can make it difficult to actually do practice that's more effective. Making practice plans but then procrastinating them or ending up "practicing" in more unhelpful ways because they're easier; feeling demotivated by failure or overwhelm about not knowing what's going wrong or what to do. Things like these have been the other biggest obstacle in translating intriguing and motivating learning theory into actual improvement in my real life. As I've gotten better with EQ skills in other parts of my life, I've found myself (building practice plans that increase my chances of) following through more, but part of me again wishes I had a more structured way of improving on this, especially with respect to learning specifically. Again, thanks for your videos, the things that I actually have implemented like free recall have been really helpful not just in my own studies, but also in tutoring friends and peers. I'm excited to see the videos coming next, and also your course!

In the three-step process, step 3 is Profit! I learned that from South Park.

I see a danger there. Making a prediction may influence equations to solve the execution -- in a bad way.

🎉😊

nice video, i would like to know your opinion abouth the ultralearning book

nice video, i would like to know your opinion abouth the ultralearning book

I might be a bit slow here, but I don't get your point at all. More precisely, what is "conceptual prediction"? What am I predicting? In what way? What is an example of a prediction? Some kind of guess what the result will be? How would that help in any way? Or does this specifically only have to do with some intuitive understanding of physical systems in physics?

Info of the particular physic book?

that works btw. i used that approach to learn physics for my exam knowing nothing about physics. its like formal logic actually in a sense that you get factually correct statements and arrange them so that output of the whole thing makes sense

would you recommend any references about the 'stop and think' trait ?

I lole how the painting behind is posing like "Oh. Damn! He is making a lot of sense"

This is awesome. Very applicable also for coding. In code reviews in my school I involuntary throw out ideas of what code does, when I read it and it sometimes annoys the evaluated, because they want to explain their code. I always ask them, to give me some time to engage with their code, before they explain it. And I make rather good progress.

you are the only one that can entertain me on this app

Commenting for the algorithm, keep up the good work!

I always ask my physics students to imagine the scenario they are working on playing out in their mind's eye as the first step. And to imagine the various parameters have sliders attached to them that one can adjust to get varying results. Really just to get them thinking about causal relationships and the parameter space in some way first, since that is what they are gonna be exploring as they solve the problem anyhow.

This is gold! Thank you!

You are a real saviour, could you make a video dicussing how "geniuses" become genus"

thank you so much, the hard solving problem where the prediction depending on time it's really hard to find which steps should to combine to get solution

Damnnn i loooooove ya video

Thank you sir, your videos are definitely helpful.