I really enjoyed this conversation with Grant. Here's the outline: 0:00 - Introduction 5:13 - Richard Feynman 9:41 - Learning deeply vs broadly 13:56 - Telling a story with visualizations 18:43 - Topology 23:52 - Intuition about exponential growth 32:28 - Elon Musk's exponential view of the world 40:09 - SpaceX and space exploration 45:28 - Origins of the Internet 49:50 - Does teaching on RUclips get lonely? 54:31 - Daily routine 1:00:20 - Social media 1:10:38 - Online education in a time of COVID 1:27:03 - Joe Rogan moving to Spotify 1:32:09 - Neural networks 1:38:30 - GPT-3 1:46:52 - Manim 1:51:01 - Python 1:56:21 - Theory of everything 2:03:53 - Meaning of life
If you point to the moon, and focus to closely you will miss all the heavenly glory. If you think your state is loneliness, you miss the beauty of "Self" ....self critical, analytical, purposeful, and self interest to your imagination.
The Feynman Effect is really the phenomenon where people mistake familiarity with understanding. It is very easy to fool yourself into thinking you understand something when you only have familiarity with the subject. This is why sometimes you study hard for a test, go over your notes without doing many examples, and end up doing poorly on an exam.
That's how studying for Physics exam become. You end up studying well enough until the exam comes and your grade is way below what you'd expect. You think that you've studied everything correctly but not for the exam, might enjoy studying physics but not enough to score high on the test. Studying for physics is not only about opening the textbook and reading that won't actually help you on the exam. Physics is about looking at every angle to ensure you understand the subject.
I think it's the other way around. Feyman gives you a really good understanding of the subject, but retention doesn't relate only to understanding, but familiriarity. I could tell you a whole bunch of simple easy to understand facts, that doesn't mean you'll remember all of the in a few months if you don't make an active effort to engrain the understanding you already have for long term use.
Grant is so eloquent. I think it's one of the reasons his videos open up new paths to understanding age old concepts. His ability to use the right word at just the right place.
Grant is a genuine hero for me. And I don't have any heroes really. The way he understands and talks about maths and its different aspects is different to anyone I can think of. I could listen to him all day. He has such a depth of understanding of not just maths but basic psychology and philosophy. He's the antithesis of so many useless professors out there who just can't communicate ideas. Genuinely great man.
To be honest, I think, as Grant said himself, he doesn't have that increadible understanding of math as other people seem to think. The problem is, people that are saying this are not experts in that field, so to them it certainly must feel so. But if you went to college, taken advanced classes of math, you would understand what I'm talking about. I'm a physics student, and I have GPA 4.0, study extremely hard, work extremely hard and yet, I've barely scratched the surface of math and physics. It is unfortunatelly unimaginable for common folk to understand how much smarter is say, Isaac Newton, Paul Dirac than even the smartest guy you personally know. And researchers are supposed to be only level bellow, which just shows how insanely smart they are compared to us "normal" people. So Grant isn't being humble, at some point in your life, you'll understand that you are not the main character.
Grant changed my life by showing me that I'm not too dumb for math, but University failed me and I had to take it into my own hands to figure it out. I hope the future of education belongs to people like Grant.
@@edwardzachary1426I believe there is a cap to muscle size, someone like a Mr Olympia such as Arnold would be a good example of the human limit to bicep size.
I think that a good solution to writers block is to relax and just write whatever comes to mind, as a brainstorm or mind map. Don't censor yourself at all, write it no matter how rubbish it seems. "No bad ideas!" Or you could choose to write down only the positive ideas, if excessive negativity is a problem. Then your mind will rapidly settle down from random chaos to some interesting topic that you've stumbled upon, which you can continue to write about in normal prose.
Grant is a major inspiration for me starting my own online math curriculum and video series and returning to earn a masters in education. Still working on lift off, this podcast gives me even more boost to my thrusters
Dude! My two all-time favorite RUclips Channels in one! Most unexpected. Most awesome!! Like the famous equation says, Lex Fridman + 3Blue1Brown = Total Awesomeness. I am so happy tonight now. Thank you!
@@JOOOOOOOE Joe is actually a lot smarter then people give him credit for. He interviews some of the greatest minds about hardest and deepest topics. Listen to some of these discussions and you'll see he's actually a verry intelligent person
@@xKhfan213x I meant from a technical perspective, should have specified more my bad. Check out his interview with Nick Bolstrom, he couldn't ask the right questions and proper continuations because tech just isn't his domain.
@@JOOOOOOOE I'll have to check that one out but it could be Nick's personality. Joe has stated that sometimes he finds it hard to connect with some guests like elon for example. Clearly joe has some tech knowledge or he wouldn't be able to hold a conversation with him that and with this podcast (elon's to be specific) he admitted he had a verry hard time with that interview. Before the cameras were rolling he said elon is this crazy person but once he gets on camera he locks himself up and it's hard to hit topics the right way, which he stated was the reason he brought the alcohol out (and possibly the pot. I'd like to see a stoned elon). He just may not have had that connection with this guest and couldn't find his rythm
Your best days lie yet in the future my friend... Educate and treat yourself and others well and find me on the other side for a cup of whatever you like... ;)
An awesome thing will be to combine all the "meaning of life" clips from all your podcasts into one video and keep it updated. Same for the "Advice for young people".
I have a little web project which can play segments of different videos seamlessly among other things, I call it mashup player. It could be useful for such a purpose.
Really enjoyed this conversation. The section on working alone really resonated with me. As a solo entrepreneur, listening to these conversations throughout my day makes me feel like I'm working at a place like Bell Labs.
hey Lex, I was recently introduced to your podcast. I’ve been going through your back catalog. I think you’re conversations are something special. Thanks for doing what you do!
Great podcast as usual. The only thing missing from these youtube uploads is the reflective and thought-provoking guitar intro/outro. I feel so incomplete without that.
All of your interviews reveal fascinating sides of very talented and insightful humanists like you. Your shows don’t go too deep into technicalities. Instead you put the spotlight on why they really enjoy their work. I love that you let them tell the many stories that brought them to where they are, and how it opened their eyes to a bigger picture. Thanks!
Entertaining while learning is very important in my opinion, inspires me to keep going when i like my prof, my brain refuses information once bored. Doing weekly flashcard review until the knowledge is longterm is my style
Richard Feynman is my hero. But that letter from Richard to his wife, specifically the comment - "I write it because it makes me warm all over inside to write to you" .... It's classic Richard, and it's what made him so deeply in tune with the surroundings. But equally, it's insular - thinking that how he feels is more important than the pure sincerity of the symbolic action he was undertaking. It's Richard through and through, if he wasn't as such - we would still be a decade or two behind.
This is such an healing podcast. Lex is curing us at such a generous and fundamental way using technology and ideas. I’d like to hereby romanticize this podcast lol
Grant saying he doesn't know enough of anything to make a significant contribution to anything had me laughing out loud. Very few people will ever make such a significant contribution to anything!
13:34 - "[In maths] everything is either trivial or impossible and its a shockingly thin line between the two, where you can find something that's totally impenetrable and then after you get a feel for it that whole, that whole subject is actually trivial in some way. So, maybe that's what goes on. Every researcher is just on the other end of that hump and it feels like it's so far away but one step actually gets them there"
As well as the teachers and professors becoming content creators, we need a sort of group of middle-men who are great at making content get together with them. Kinda like numberphile etc. I think it's a very underrated concept but (IMO obviously) that would a superb move for any uni to have their own content production team focus on actual educational videos. Seeing all the personalities and yours and others videos makes whatever institution they represent much more interesting.
1:35:37 “The reason that gradient descent works well with neural networks and not just choose how you want to parameterize the space... Is that that layered structure allows you to decompose the derivative in a way that makes it computationally feasible.” That’s a doozy of a sentence.
What an excellent interview. On the topic of MANIM, hands down the best tool I have used as a teacher. It is hyper engaging and is a resource of the future! Any educators who see this, lets collaborate these resources?
Ok I do not leave comments and sometime I do not even hit the like button because I do not think about it. Thanks a lot for your work both of you, I love your videos Lex, I love Grant's videos, while I relate to Lex's unhappiness coming from too much geekery, when I spend too much time on the internet but it's for watching some of your work guys If I'm doing some introspection I feel kind of happier.
adamgm84 I’m actually curious as to the percentage of dislikes can be attributed to misclicks. Like if you’ve seen his ted talk, the question is what person who can dislike that?
It has happened to me, where I found that a video a really liked years ago and I found there is a dislike on it. So yeah, I agree that is what should be happening.
I would love to see you interview : (1) Lawrence Krauss (2) Jordon Peterson (3) Adam Neely( Musician) .... yes... I know this one is different, but it would be an amazing discussion for sure....❤️ @itsadamneely (4) Joscha Bach( a thousand more times...)❤️ Love your work Dr. Fridman....... You inspire me everyday in my graduate studies...❤️🙏🏻
I've been dealing with this, Amblyopia. If you cover the strong one and do something like reading, it gets better. I do it like an hour a night, when I remember. It's gotten better...
As someone who is just self studying algebraic topology, I mostly agree with you! People think topology is some obscure and niche subject, but it is actually incredibly incredibly powerful and beautiful. It should be divided and is usually divided into two subjects in the professional world. One is general topology and the other is algebraic topology. The subject itself revolutionized mathematics on several fronts! Category theory, torsion in algebra, schemes and sheaves, covering theory and manifold theory lead to powerful things in analysis(Riemann surfaces for example), etc. General topology had the goal of abstracting away the ideas of Cauchy and Weierstrass and build from there. How can we generalize continuity? How about the notion of sth being in the neighborhood of another thing? Yes it is easy when you visualize points in an euclidean space, but how about functions? That has been dealt with in analysis. But how about groups? This is getting weird isn't it. You see, you can have a group(look it up, grant just made a good video about his) act on a geometric shape, like a torus, and you get a new object. How do you understand whats going on there? Then you have weird things like topologies on ordinal numbers, Zariski topology, you can even start talking about topology in context of logic(You can actually use topology to give a visual and geometric "representation" for Kripke's worlds in his semantics of modal logics. Look up Topology and First Order Modal Logic written by Mkoconnor on xorshammer.com. Shouldn't be too difficult if you have some math background) I HIGHLY recommend watching Tadashi Tokieda's lectures on Topology and Geometry. He is one of the best mathematics educators I have ever seen and he gets the point across, trust me. These don't really touch algebraic topology I don't think, but you get a closer look at what is happening. If you want to see algebraic topology in action, then you have the wonderful textbook by Allen Hatcher and Piere Albin recorded great lectures as he teaches from this book in class. I recommend this straight away, because it is too difficult to present any ideas from algebraic topology to a lay person just like that. The subject is too heavilly rooted in parts of math that lay people simply know nothing about, especially algebra and general topology are both heavilly required. This is why you will rarely ever see an actual layman topology video that is about the "good stuff" mathematicians care about. Finally, I do disagree with a point you made. You said it is not needed to go on this long road of general topology to reach AT, but I disagree to an extend. You absolutely NEED to go through a semester of general topology and even that it should be pretty fast paced and advanced. A full year course on general topology that covers everything from the basics to topological manifolds and CW complexes would be ideal. You can skip analysis stuff for the most part, like uniform spaces and most of compact open topology will be too detached from AT(learning the required basics of compact open topology from Hatcher's appendix in AT should be enough). Well idealy you would need that also, because often you need some non trivial thing from compact open topology to construct say a retraction of a space onto a subspace. But you NEED to master the basic definitions. The topics of compactness, connectedness and quotient spaces are the most essential to absolutely master! You cannot understand ANY algebraic topology before you have a good grasp of these things. And "a good grasp" is putting it mildly. AT is where you take the training wheels of cutesey examples off and go for the deep end. Examples that push your visualisation and topology knowledge to the limit are not rare(for those of you who know, the Torus Knot example for Van Kampen Theorem stalled my progress for 2 days or so. Hatcher is genius in forcing you to do the god damn work and that he did here). At least Hatcher tends to do that often and I LOVE it.
2:03:19 GS: "Build a habit of feeling what it's actually like to come to resolution" LF: "As opposed to-and it can be enjoyable-in awe of the fact that you don't understand anything" GS: "Maybe people get entertainment out of that, but you won't grow"
I must say that whenever I watch your (3b1b's) videos, I not only get that "intellectual itch", but also retain the main ideas behind the lessons. I think it could be because I am a few years younger than an undergrad, as a result, it might be stored in some other memory compartment close to childhood memories? Regardless, your videos are extremely inciteful, always. I have sympathy for those who came before me without the possibility to watch your videos and envy those who will come after me with access to plenty more of your videos.
When you said youtube might not be forever it really hit me. Someone needs to make a platform designed to be permanent. Once you upload, it cannot be deleted and will stay for generations to come.
A note on what Grant says at 1:35:32 : There's nothing special about the layer structure. You can compute the gradient of pretty much any function efficiently, using reverse-mode automatic differentiation. This is really the same thing as backpropagation, but it's clear that it works for nearly arbitrary functions. In short, evaluating the gradient of a function takes about as much effort as computing the function in the first place.
When I get invitation from Lex Fridman podcast maybe someday in the future, I will know that I've done enough influential and important things in my life in order to be qualified for this
1:21:10 Plasma Physics Channel with Prof. Alexander Fridman would be awesome idea. With latest developments of ITER it's getting really hot (pun intended!) topic.
I really enjoyed this conversation with Grant. Here's the outline:
0:00 - Introduction
5:13 - Richard Feynman
9:41 - Learning deeply vs broadly
13:56 - Telling a story with visualizations
18:43 - Topology
23:52 - Intuition about exponential growth
32:28 - Elon Musk's exponential view of the world
40:09 - SpaceX and space exploration
45:28 - Origins of the Internet
49:50 - Does teaching on RUclips get lonely?
54:31 - Daily routine
1:00:20 - Social media
1:10:38 - Online education in a time of COVID
1:27:03 - Joe Rogan moving to Spotify
1:32:09 - Neural networks
1:38:30 - GPT-3
1:46:52 - Manim
1:51:01 - Python
1:56:21 - Theory of everything
2:03:53 - Meaning of life
I love how you end up talking about the meaning of life
How in the world are you able to keep creating incredible content?? Thanks, keep up the good work
how about the guy from the RUclips channel minutephysics then? or kurzgesagt
If you point to the moon, and focus to closely you will miss all the heavenly glory. If you think your state is loneliness, you miss the beauty of "Self" ....self critical, analytical, purposeful, and self interest to your imagination.
Fantastic, had the privilege of meeting Grant after a few back and forth emails after his bitcoin video. Hope to see him in person soon.
What an iconic voice.
He has not taken his sleep since long time for reason of doing maths homework
The Feynman Effect is really the phenomenon where people mistake familiarity with understanding. It is very easy to fool yourself into thinking you understand something when you only have familiarity with the subject. This is why sometimes you study hard for a test, go over your notes without doing many examples, and end up doing poorly on an exam.
That's how studying for Physics exam become. You end up studying well enough until the exam comes and your grade is way below what you'd expect. You think that you've studied everything correctly but not for the exam, might enjoy studying physics but not enough to score high on the test. Studying for physics is not only about opening the textbook and reading that won't actually help you on the exam. Physics is about looking at every angle to ensure you understand the subject.
indeed, which is why doing practice questions is so important
I think it's the other way around.
Feyman gives you a really good understanding of the subject, but retention doesn't relate only to understanding, but familiriarity. I could tell you a whole bunch of simple easy to understand facts, that doesn't mean you'll remember all of the in a few months if you don't make an active effort to engrain the understanding you already have for long term use.
Yes!
@@tarek3735 yeah, I'm rly feeling this rn
Grant is so eloquent. I think it's one of the reasons his videos open up new paths to understanding age old concepts. His ability to use the right word at just the right place.
He said in this talk that the most time consuming part of his video-making work is writing script.
his other podcast he said he practiced this skill he wasn't always good at speaking, which is inspiring
and if not scripted, that word is "like"
So true
Yes, he's amazingly articulate. On a scale of 1 to 10 on articulation he's a 10. I wish I were half as articulate.
Grant is a genuine hero for me. And I don't have any heroes really. The way he understands and talks about maths and its different aspects is different to anyone I can think of. I could listen to him all day. He has such a depth of understanding of not just maths but basic psychology and philosophy. He's the antithesis of so many useless professors out there who just can't communicate ideas. Genuinely great man.
💯
To be honest, I think, as Grant said himself, he doesn't have that increadible understanding of math as other people seem to think. The problem is, people that are saying this are not experts in that field, so to them it certainly must feel so. But if you went to college, taken advanced classes of math, you would understand what I'm talking about. I'm a physics student, and I have GPA 4.0, study extremely hard, work extremely hard and yet, I've barely scratched the surface of math and physics. It is unfortunatelly unimaginable for common folk to understand how much smarter is say, Isaac Newton, Paul Dirac than even the smartest guy you personally know. And researchers are supposed to be only level bellow, which just shows how insanely smart they are compared to us "normal" people. So Grant isn't being humble, at some point in your life, you'll understand that you are not the main character.
Dear Lex, I hardly ever comment on RUclips video's, but I think you're doing a fantastic job. Thank you very much for everything you create.
3yrs later and this is still one of the best conversations on the internet. Thank you Lex for all that you do for the human race
Grants been going gym. he'll finish your homework and steal your girl
I for one wouldn't mind
I think those arms are due to hours of coding
@@ettvanligtkonto wife
Dude, Grant is absolutely SEXY
How can deado steal any girl?
Lex, Can you interveiw Sal Khan?
Yes pleaseeee.
This is a must
PLEASE
Hell yeah!
He said he doesnt read comments haha. But it would be cool
Grant changed my life by showing me that I'm not too dumb for math, but University failed me and I had to take it into my own hands to figure it out. I hope the future of education belongs to people like Grant.
exponential biceps
Can I bicep be bigger than the length of shoulder to elbow? Limiting factor?
@@edwardzachary1426 Since my last comment, I spent 1 year trying to grow my biceps and I can assure it did not follow an exponential growth
@@edwardzachary1426I believe there is a cap to muscle size, someone like a Mr Olympia such as Arnold would be a good example of the human limit to bicep size.
"The solution to writer's block is to read"
Gotta get that printed on a t-shirt.
I think that a good solution to writers block is to relax and just write whatever comes to mind, as a brainstorm or mind map. Don't censor yourself at all, write it no matter how rubbish it seems. "No bad ideas!" Or you could choose to write down only the positive ideas, if excessive negativity is a problem. Then your mind will rapidly settle down from random chaos to some interesting topic that you've stumbled upon, which you can continue to write about in normal prose.
Grant is so eloquent, so bright and sunny. Thank you for the interview.
Grant is a major inspiration for me starting my own online math curriculum and video series and returning to earn a masters in education. Still working on lift off, this podcast gives me even more boost to my thrusters
Damn! Grant is getting buff!
Perfect timing. I just finished watching your first interview with Grant and here you are uploading a second one
Dude! My two all-time favorite RUclips Channels in one! Most unexpected. Most awesome!!
Like the famous equation says, Lex Fridman + 3Blue1Brown = Total Awesomeness. I am so happy tonight now. Thank you!
Very great discussion! Thanks for doing it! I was listen to it while scanning/digitising all my old Math school notebooks :)
heyyy! brightside of math!
I would like 2 things:
- Attached recording date (just curious)
- Make someone else interview yourself. We wanna hear you as well
@@kurtisstreutker6196 joe is not technically literate and a discussion with him is mostly superficial, albeit entertaining.
@@JOOOOOOOE Joe is actually a lot smarter then people give him credit for. He interviews some of the greatest minds about hardest and deepest topics. Listen to some of these discussions and you'll see he's actually a verry intelligent person
@@xKhfan213x yeah, Rogan is not some superficial idiot, those memes need to die
@@xKhfan213x I meant from a technical perspective, should have specified more my bad. Check out his interview with Nick Bolstrom, he couldn't ask the right questions and proper continuations because tech just isn't his domain.
@@JOOOOOOOE I'll have to check that one out but it could be Nick's personality. Joe has stated that sometimes he finds it hard to connect with some guests like elon for example. Clearly joe has some tech knowledge or he wouldn't be able to hold a conversation with him that and with this podcast (elon's to be specific) he admitted he had a verry hard time with that interview. Before the cameras were rolling he said elon is this crazy person but once he gets on camera he locks himself up and it's hard to hit topics the right way, which he stated was the reason he brought the alcohol out (and possibly the pot. I'd like to see a stoned elon). He just may not have had that connection with this guest and couldn't find his rythm
I’m always so exited to see a pure mathematician on your podcast. This man helped me tremendously through my schooling.
As always, thank you Lex for your desire to understand, which permeates through RUclips and connects with all of us listeners. Have a good one!
Thank you soooooo much for putting this video out. I'm going through a tough time and this is great for my mental health. Thank you both so much!
Your best days lie yet in the future my friend...
Educate and treat yourself and others well and find me on the other side for a cup of whatever you like... ;)
Grant you made me love mathematics. You are a true educator
Grant starts talking and instantly i'm fascinated, that's how powerful his videos have been for me. :D
Lex and Grant, this is one of the best conversation I’ve seen this year. Thanks a lot.
An awesome thing will be to combine all the "meaning of life" clips from all your podcasts into one video and keep it updated. Same for the "Advice for young people".
I have a little web project which can play segments of different videos seamlessly among other things, I call it mashup player. It could be useful for such a purpose.
Thank you again! Your amazing guests and your skill as an interviewer add up to a thought-provoking and worthwhile time for me.
Really enjoyed this conversation. The section on working alone really resonated with me. As a solo entrepreneur, listening to these conversations throughout my day makes me feel like I'm working at a place like Bell Labs.
1:01:00 I can relate, but fyi, so far spending time listening to your podcast always yielded more happy and productive version of me, thanks Lex!
This is an incredible interview and so wrought with an authenticity I just haven't found anywhere else on the internet. Thanks!
I love 3 blue 1 brown! His videos helped me so much with linear algebra. I just graduated with a math and engineering degree! Thank you so much
Grant be like, "ayy imma find a local maximum of the vascularity function before this podcast wit lex"
Lex, I love your podcasts. You're a great interviewer and have excellent guests. Also inspired me to teach myself computer science!
Same! Discussions with programming language designers made me go and figure out how they work.
Agent Smith: "Mr. [S]anderson. Welcome back, we missed you." - I'm a cornball lol.
hey Lex, I was recently introduced to your podcast. I’ve been going through your back catalog. I think you’re conversations are something special. Thanks for doing what you do!
I think after watching this my IQ points increased from 90 to 91
we are in the logarithmic side of the curve
Soon the whole pond my man
Watch it 100 times and you’ll be a genius at that rate
Great podcast as usual. The only thing missing from these youtube uploads is the reflective and thought-provoking guitar intro/outro. I feel so incomplete without that.
All of your interviews reveal fascinating sides of very talented and insightful humanists like you. Your shows don’t go too deep into technicalities. Instead you put the spotlight on why they really enjoy their work. I love that you let them tell the many stories that brought them to where they are, and how it opened their eyes to a bigger picture. Thanks!
Entertaining while learning is very important in my opinion, inspires me to keep going when i like my prof, my brain refuses information once bored. Doing weekly flashcard review until the knowledge is longterm is my style
1:52:20 GLSL shaders are amazing, glad he's implemented them into Manim
I’ve never clicked on a video faster. Two of my favourite people talking again!!
57:55 We're back to *"the storm of the Russian soul"* vs. *"California sunshine"*
california sunshine omg i love that nickname for grant
Richard Feynman is my hero. But that letter from Richard to his wife, specifically the comment - "I write it because it makes me warm all over inside to write to you" .... It's classic Richard, and it's what made him so deeply in tune with the surroundings. But equally, it's insular - thinking that how he feels is more important than the pure sincerity of the symbolic action he was undertaking. It's Richard through and through, if he wasn't as such - we would still be a decade or two behind.
Can't believe i listened to the entire 2 hour podcast !
everytime i see grant in a podcast , it just lights up my day
The end was the icing on the cake. Beautiful!
Grant, you definitely should do more videos on topology. I am having the problem you mention of connecting the two perspectives about topology.
Feynman letter to lost wife is so redolent of a Yiddish sensibility, full of ironic love. Thanks for sharing.
One of the best conversations I’ve had the privilege of listening to.
This is such an healing podcast. Lex is curing us at such a generous and fundamental way using technology and ideas. I’d like to hereby romanticize this podcast lol
I'm addicted to 3b1b, he's has changed how linear algebra is studied worldwide, always recommended by my professors and online courses on coursera ❤️
Grant saying he doesn't know enough of anything to make a significant contribution to anything had me laughing out loud.
Very few people will ever make such a significant contribution to anything!
Thank you Lex for all your videos! I am learning a lot through them, and they make me Google some discussed topics to learn more!
13:34 - "[In maths] everything is either trivial or impossible and its a shockingly thin line between the two, where you can find something that's totally impenetrable and then after you get a feel for it that whole, that whole subject is actually trivial in some way. So, maybe that's what goes on. Every researcher is just on the other end of that hump and it feels like it's so far away but one step actually gets them there"
Really enjoy Grant opinions on theses topics, Lex did a great job.
This is a nice meaty interview with some interesting points about education, creativity, etc... I'll definitely have to return to it again.
I am not sure if we deserve this combination. It was a blissful two hours ❤️
Both of you add great value to the internet. Thanks.
Since you do read the comments: congratulations! I really enjoy your podcast, and I love Grant’s channel
Lol the ads were fun and pretty informative 😂
I hardly comment but super happy both of your channels exist!
As well as the teachers and professors becoming content creators, we need a sort of group of middle-men who are great at making content get together with them. Kinda like numberphile etc. I think it's a very underrated concept but (IMO obviously) that would a superb move for any uni to have their own content production team focus on actual educational videos. Seeing all the personalities and yours and others videos makes whatever institution they represent much more interesting.
This one deserves the engagement!
1:35:37
“The reason that gradient descent works well with neural networks and not just choose how you want to parameterize the space... Is that that layered structure allows you to decompose the derivative in a way that makes it computationally feasible.”
That’s a doozy of a sentence.
What an excellent interview. On the topic of MANIM, hands down the best tool I have used as a teacher. It is hyper engaging and is a resource of the future! Any educators who see this, lets collaborate these resources?
1.27 million. You deserve this Lex.
Lex "play devil's advocate" Fridman :D This is my favourite podcast atm. So many great quests, and great host!
Ok I do not leave comments and sometime I do not even hit the like button because I do not think about it. Thanks a lot for your work both of you, I love your videos Lex, I love Grant's videos, while I relate to Lex's unhappiness coming from too much geekery, when I spend too much time on the internet but it's for watching some of your work guys If I'm doing some introspection I feel kind of happier.
lex telling me about shaving is something I never knew I needed
1:38:47 Epic in hindsight!
🙏 absolutely great as usual. Grant changes lives for the better.
Vaguely aware of 3Blue1Brown, but listening to this and Interview#1, I love this guy!
Absolutely love 3blue1 brown. So excited to listen to this.
Wonderful Conversation, Thank you Lex
1 person was so excited to like that they hit the wrong button.
adamgm84 I’m actually curious as to the percentage of dislikes can be attributed to misclicks. Like if you’ve seen his ted talk, the question is what person who can dislike that?
It has happened to me, where I found that a video a really liked years ago and I found there is a dislike on it. So yeah, I agree that is what should be happening.
25 of them now
I would love to see you interview :
(1) Lawrence Krauss
(2) Jordon Peterson
(3) Adam Neely( Musician) .... yes... I know this one is different, but it would be an amazing discussion for sure....❤️ @itsadamneely
(4) Joscha Bach( a thousand more times...)❤️
Love your work Dr. Fridman....... You inspire me everyday in my graduate studies...❤️🙏🏻
His videos are awesome! Keep on keeping on Grant! 👍
I always love your videos, from content to charisma, you spark my interest on this topics.
10:00
Have good relationship with the professor for them to be your mentors.
I've been dealing with this, Amblyopia. If you cover the strong one and do something like reading, it gets better. I do it like an hour a night, when I remember. It's gotten better...
As someone who is just self studying algebraic topology, I mostly agree with you! People think topology is some obscure and niche subject, but it is actually incredibly incredibly powerful and beautiful. It should be divided and is usually divided into two subjects in the professional world. One is general topology and the other is algebraic topology. The subject itself revolutionized mathematics on several fronts! Category theory, torsion in algebra, schemes and sheaves, covering theory and manifold theory lead to powerful things in analysis(Riemann surfaces for example), etc.
General topology had the goal of abstracting away the ideas of Cauchy and Weierstrass and build from there. How can we generalize continuity? How about the notion of sth being in the neighborhood of another thing? Yes it is easy when you visualize points in an euclidean space, but how about functions? That has been dealt with in analysis. But how about groups? This is getting weird isn't it. You see, you can have a group(look it up, grant just made a good video about his) act on a geometric shape, like a torus, and you get a new object. How do you understand whats going on there? Then you have weird things like topologies on ordinal numbers, Zariski topology, you can even start talking about topology in context of logic(You can actually use topology to give a visual and geometric "representation" for Kripke's worlds in his semantics of modal logics. Look up Topology and First Order Modal Logic written by Mkoconnor on xorshammer.com. Shouldn't be too difficult if you have some math background)
I HIGHLY recommend watching Tadashi Tokieda's lectures on Topology and Geometry. He is one of the best mathematics educators I have ever seen and he gets the point across, trust me. These don't really touch algebraic topology I don't think, but you get a closer look at what is happening.
If you want to see algebraic topology in action, then you have the wonderful textbook by Allen Hatcher and Piere Albin recorded great lectures as he teaches from this book in class. I recommend this straight away, because it is too difficult to present any ideas from algebraic topology to a lay person just like that. The subject is too heavilly rooted in parts of math that lay people simply know nothing about, especially algebra and general topology are both heavilly required. This is why you will rarely ever see an actual layman topology video that is about the "good stuff" mathematicians care about.
Finally, I do disagree with a point you made. You said it is not needed to go on this long road of general topology to reach AT, but I disagree to an extend. You absolutely NEED to go through a semester of general topology and even that it should be pretty fast paced and advanced. A full year course on general topology that covers everything from the basics to topological manifolds and CW complexes would be ideal. You can skip analysis stuff for the most part, like uniform spaces and most of compact open topology will be too detached from AT(learning the required basics of compact open topology from Hatcher's appendix in AT should be enough). Well idealy you would need that also, because often you need some non trivial thing from compact open topology to construct say a retraction of a space onto a subspace. But you NEED to master the basic definitions. The topics of compactness, connectedness and quotient spaces are the most essential to absolutely master! You cannot understand ANY algebraic topology before you have a good grasp of these things. And "a good grasp" is putting it mildly. AT is where you take the training wheels of cutesey examples off and go for the deep end. Examples that push your visualisation and topology knowledge to the limit are not rare(for those of you who know, the Torus Knot example for Van Kampen Theorem stalled my progress for 2 days or so. Hatcher is genius in forcing you to do the god damn work and that he did here). At least Hatcher tends to do that often and I LOVE it.
2:03:19
GS: "Build a habit of feeling what it's actually like to come to resolution"
LF: "As opposed to-and it can be enjoyable-in awe of the fact that you don't understand anything"
GS: "Maybe people get entertainment out of that, but you won't grow"
So underrated!
Great video! A post from some one who doesn't normally post.
I must say that whenever I watch your (3b1b's) videos, I not only get that "intellectual itch", but also retain the main ideas behind the lessons. I think it could be because I am a few years younger than an undergrad, as a result, it might be stored in some other memory compartment close to childhood memories? Regardless, your videos are extremely inciteful, always. I have sympathy for those who came before me without the possibility to watch your videos and envy those who will come after me with access to plenty more of your videos.
Very well said, and a good point, too!
Two of my most favorite youtuber at the same time.
Lex, I love you man. Keep up the Quines
Gold: 27:16 ~ 27:36
Thanks a lot for this lex!
This was an amazing interview, thank you.
What I've learnt from this is, if you put 2+ talented, maths-friendly content creators in a room together, the perimetre is effectively inifinite x
What an incredible conversation
Feels good to tune in earlier i m officially the second to watch it after lex
When you said youtube might not be forever it really hit me. Someone needs to make a platform designed to be permanent. Once you upload, it cannot be deleted and will stay for generations to come.
"Numberphile traps teacher(s) in a room ..." to make them "work" 🤣🤣🤣
Grant voice is so soothing
A note on what Grant says at 1:35:32 : There's nothing special about the layer structure. You can compute the gradient of pretty much any function efficiently, using reverse-mode automatic differentiation. This is really the same thing as backpropagation, but it's clear that it works for nearly arbitrary functions. In short, evaluating the gradient of a function takes about as much effort as computing the function in the first place.
Great pod. Learnt 200 new words
This podcast became my new favorite.
When I get invitation from Lex Fridman podcast maybe someday in the future, I will know that I've done enough influential and important things in my life in order to be qualified for this
1:21:10 Plasma Physics Channel with Prof. Alexander Fridman would be awesome idea. With latest developments of ITER it's getting really hot (pun intended!) topic.