Dear Dr. Peyam, THANK YOU !! And engineering we’re taught how to use convolution, but never learn where the hell it comes from. Your explanations are like a brain massage 💆♂️. Thank you, thank you! You know an explanation is good when it not only answers a question I hadn’t even thought of, but also opens my mind to other ways of thinking about math. So much fun! Danka!!
General comment: Convolution can be thought of as a measure of self-similarity. The more self similarity between and within the two functions, the larger the convolution integral’s value. (There is the group theory connection)
This intuition makes a lot of sense because for each x the convolution is essentially just an inner product between one function and another function that has been reflected and shifted.
Chobes 182 so if thé function and its shiftted values are strongly correlated (or even equal) the convolution integral approaches the integral of the square of the function. The more dissimilar the shift and the non-shifted values are, the integral can be greater or lesser than the integral of the square.
Thanks for this video. For us, statisticians, is a very important result for it is related to finding the distribution of the sum of two random variables. So, in general I wish to express our appretiation for your efforts.
Damn, this is amazing, brother. Though I'll need to watch this video like 2 or 3 more times to connect the dots. Keep up with the good work! Really, you are one of the most interesting and useful TH-cam channels I've been subscribed to
Okay, I really am seeing what you did there, but I feel like what makes this really suggestive to me is looking at the each power of x as a basis function. Wooooow, this is so much more abstracted and interesting compared to the way I usually look at it as a moving inner product
Thank you so much for the baby step explanation ! It became completly intuitive thanks to the way you ve presented. It ! We want more of awsome content
Very interesting. I wonder if we can draw other intuitions from polynomial functions and transfer them to general analytical functions. Anyway analytical function can be aproxymated locally by Taylor series. But in this case analogy seems to work not only locally but also in whole range.
I feel that this definitely helped me. Not really sure why you began discussing the continuous convolution though. The whole polynomial discussion is perfectly applicable in the context of discrete convolution. Anyway, for whatever reason, motivating with polynomial multiplication somehow did it for me. Thanks! I'm also finding it helpful in higher dimensions to think in terms of multiplying polynomials (the number of variables = the number of dimensions): To find the coefficient for x_1^n_1 * x_2^n_2, you multiply coefficients of the input polynomials where the exponents add up to n_1 and n_2. This kind of explains why you need to flip the "kernel" (in all dimensions) when you think of convolution as a "sliding dot product": when you flip the kernel, the coefficients that you need to multiply "pair up" (such that the exponents add up to n_i). Also, I really like your sanity check: the two arguments MUST sum to x! Sounds gimmicky, but I'm pretty sure that will help me to remember.
Thanks a lot doctor payam ! Convolution is really confusion topic for me ... I would like to ask that, is convolution useful for mathematicians. ..? It is part of digital signal processing as per my information
@@drpeyam Here's "the" classic video on convolution from the engineering school perspective: th-cam.com/video/_vyke3vF4Nk/w-d-xo.html you will have to forgive the "cool 70s disco" look of the professor, it was indeed the 70s, so......he looks the part, (but, Prof Oppenheim is/was the guru on signals and systems theory") This is immensely useful math. immensely.
@@sandorszabo2470 Hello. But prof Peyam is nearly the same: when he talks of convolution in terms of multiplying two polynomials, prof oppenheim talks about "linear time invariant" systems which produce polynomial sums as "outputs" of multiple inputs in the LTI context! Almost similar! But yes, the original intuition was from the Engg department side, historically.
Brother I know mechanical engineers could find resonance but when I had a deep thought on this resonance Is an slow accumulation of energy which is accumulated very high in small installments when the frequencies match if you strike a turning fork of 50 hz you get the same frequency of vibration on another tuning fork so they both vibrate if you strike it harder the amplitude changes hence loudness is a human factor the frequency is the same the languages that human speak through out the world the sound only resonate your ear drum for few seconds my question is that the harmonics is the fundamental frequency and overtones are the frequency that follow it take a word in any language you spell it according to convolution the thing scales and ques and stack the signal so convolution can be used to model resonance so when your ear drum vibrates it vibrates so the electrical signals are carried to brain like tuning fork ear drums vibrate within the audible spectrum 20 hz to 20000 hz hence resonance is caused by the word we speak and within the audible range the ear drums vibrate and we make sense of words I have seen in one videos on TH-cam that due to harmonics in any sound causes resonance which could be modelled by convolution recalling the resonance its destructive because slow and steady accumulation of sound on the mass causes high stress and high energy to build inside and stress increase and the system fractures or collapses but our ear drum hearing the sound from human languages try to vibrate but why our ear drum when subjected to continuous exposure of sound does not fracture or rupture like a wine glass iam not telling about high loud sound higher than 80 db but a audible range sound within the frequency of 20 hz to 20000 hz under continuous exposure why it's not damaging it again not failure by high energy but low one in synchronisation on air . But I tried it in my students when I told them to be quite in class they did not listen to me so I took my phone and set an frequency 14000 hz and they told it was irritating the idea of resonance is "small effort but large destruction " just like Tacoma bridge where the wind just slowly accumulated energy on the bridge and it collapsed it so my conclusion is if an audible frequency at continuous exposure to an human ear can it cause bleeding again "small effort but great destruction" sorry for the long story I you are able to reach hear you must be as curious as me so still not finished the ear drum is shook by harmonics in the sound we make by the words( or )overtones in the sound we make by the words I know harmonics is the fundamental frequency and overtones are following it which under slow and steady accumulation of sound energy resonates and could damge the ear drums again "small effort but big destruction" not to mention we assume the person is in coma or brain dead hence when the sound irritates him he or she could not make a move so my question is so simple normally human ear responds to harmonics or overtones according to convolution which could be a disaster but with minimal effort 🙏🙏🙏🙏 at here I could be wrong because harmonics can also be used to construct sound so can it be destructive or the overtones which are the trouble makers and which one according to this gives a response curve when two signals convolved by harmonics or overtones which is destructive but with minimal effort and convolution happens when ear drums oscillate is by harmonics or the overtones or also the trouble makers there
Convolution is integration of *f(τ).g(t-τ) dτ* *(τ , τ+dτ)* is an infinitesimally small time period for which we are assuming the values of *f(τ)* and *g(t-τ)* remain constant. *f(τ)* is the evolution of f(t) until the time instant *τ* . *g(t-τ)* is the version of g(t) which came into existence at the time instant *τ* .
I understand that convolution is analogous to the multiplication of two polynomials. The intuition here is to express any signal f in terms of its impulses, just like coefficients of a polynomial. It makes sense, thanks. But I still do not understand why we convolute a signal when it is filtered. We may multiply the signal with the filter point-wise.
When I transform a function f(x )into an endless series, and also the function g(x) Then I create a convolution with these two powerseries in your discrete way, with Sigma and Indices. Is the resulting series the same as I transform the continous version (f*g)(x) into a series?
That was my realization from the video too. Now that I think about it, I think that's the result from multiplication of taylor series both fixed about a point a.
Hey Dr. Peyam. I had this issue in undergrad too! Thank you for the video. Out of curiosity, for convolutional neural networks, whenever they talk about the "window" in convolving images, would the "window" be analogous to getting the coefficient of a particular degree on this example?
Remark 1: This one is pretty nice: th-cam.com/video/QmcoPYUfbJ8/w-d-xo.html ["What is convolution? This is the easiest way to understand" by Discretised] It is in terms of integration of processes with fading intensity, but it is amenable for economic interpretation as well. Remark 2: This multiplication by gathering indices that sum up to a constant is crucial for the Cauchy product of two infinite series instead of polynomials (Mertens theorem). Remark 3: This convolution is with respect to time. In image manipulation the convolution is with respect to space (a kind of weighted averaging over pixels). That "spatial convolution" in the continuous case leads to an integral transform. One of the functions under convolution is then called a kernel. Just loose thoughts.
Convolution is a term that really might better be described as “self-similarity”. It even has application to music theory! (THERE is the Group Theory connection!!! And even Biology!!!)
Hey Dr Peyam, can we perhaps see a more rigorous definition of what you mean by “continuous polynomials”, how functions can be described in terms of them, and how that leads to the convolution? I would also love to see how this connects to the view of convolution in terms of linear functionals, as Physics Videos By Eugene made an extensive video on that which at least I didn’t really understand... Anyhow, thanks a lot for this!
@@drpeyam Couldn't we define them as an integral average? like the definite integral from zero to n of a(t)*x^t dt, all of that divided by n to "cancel out" the "dt" part, if we look at it from a perspective of dimensional analysis like done in physics
My profs at The Cooper Union, 1967-1971 likes to say the variable integrated over is “integrated out”..... which I hold is not accurate, as it is only in appearance, vanished..... the functions evaluated at each point of the “integrated out” variable contribute to the sum, as well as the end points. As the variable EXPLICITLY vanishes, it “goes away. By the way, Dr. Tabrizian, what is “f hat” you refer to in the Fourier transform description of the convolution? Please explain?
It tells to integrate wrt y and keep x fixed (the resulting function is of x variable). Integration wrt y is a continuous analog of summation over the index (also termed y at the end of the presentation, to highlight the jump from a discrete to the continuous case).
It made sense to me in some way...I still want to know the advantages of 'reflecting' and 'shifting' a function and then multiplying that with another function. If we do not 'reflect' then what ? Shifting I can understand..we have to keep moving window everywhere..
Now I'm really curious if this interpretation could be used to give a more intuitive interpretation of volterra series analysis. Which is my favorite analysis technique that I learned in electrical engineering
Wikipedia says the symbol from your thumbnail means something called ,,cross corelation'' and it's simmilar to convolution. I hope somewhere in future you will make a video about that.
That means you could get the taylor series of sin^2x, that would be useful in solving diff equations by solving for a taylor series. You could also continue value for sinx in the infinities.
Oh so i finally understood. F and g aren't time functions. They are the formula for the element of the taylor series. Sinx isn't f. F is i^(k-1)*(1/k!)* ( k mod 2)
I could never find an explanation of why (graphically) you have to reflect one function, multiply both and integrate. I see its too keep the indices to always sum the same?
do alternate notations for convolution exist. I hate that notation for convolution since i love using * to mean multiplicaiton and do so quite frequently.
It was amazing but at the moment you replaced coefficient with the function itself, I didn't understand actually how you did this. Is there anyone who can make it clear for me? Thanks.
You were right to both hesitate and then ignore the possibility that you had misspelled “coefficients”. English is difficult because it is FULL of irregularities.... this is one instance of a violation of the rhyme “I before E (edited 12-12-2023) except after C or when sounding like “Eh” (“long” A) as in Neighbor and Weigh”. Had you bothered to worry about that during the lecture, it would have impeded progress and the continuity (smile) of the discussion.
Idk if this helped with my intuition, but it did kinda blow my mind.
instaBlaster
The result is not important , whats important is the process.. - Dr. Peyam..... sir you are extremely motivational to me
OMG, this is the simplest explanation of convolution I've ever come across, thank you so much!!!
This arguably might be the most important video on youtube. I wanted to cry at the end from an epiphany.
Thanks so much 🥹🥹
Dear Dr. Peyam, THANK YOU !! And engineering we’re taught how to use convolution, but never learn where the hell it comes from. Your explanations are like a brain massage 💆♂️. Thank you, thank you! You know an explanation is good when it not only answers a question I hadn’t even thought of, but also opens my mind to other ways of thinking about math. So much fun! Danka!!
General comment: Convolution can be thought of as a measure of self-similarity. The more self similarity between and within the two functions, the larger the convolution integral’s value. (There is the group theory connection)
Interesting!
Dr Peyam yes! That is why & how it applies to biology and music theory!!
Wow! I didn't know about that! Very very cool!
This intuition makes a lot of sense because for each x the convolution is essentially just an inner product between one function and another function that has been reflected and shifted.
Chobes 182 so if thé function and its shiftted values are strongly correlated (or even equal) the convolution integral approaches the integral of the square of the function. The more dissimilar the shift and the non-shifted values are, the integral can be greater or lesser than the integral of the square.
Thanks for this video. For us, statisticians, is a very important result for it is related to finding the distribution of the sum of two random variables. So, in general I wish to express our appretiation for your efforts.
I love your attitude, sir! I'm motivated just hearing you speak, let alone how good you explain the subject.
Brilliant explanation! Brilliant - makes it look so natural and so simple. Thanks heaps. I had been really curious about where it came from.
Convolution is for most mortals, a mathematical nightmare and absolutely non intuitive. But this explanation, makes it more obvious. So thumbs up !!!
One word in my mouth : Great ! Bravo !
That clarified convolution once and for all 💯💯
I loved this interpretation. Thank you, Dr. Peyam.
Thank you! The students you teach are very lucky :) And we are lucky to be able to watch your videos
Thank you. Finally understood the intuition behind this pop operation called convolution.
Thanks Doc, easy to follow. This is a good generalisation of convolution.
This is wonderful. It does really make sense.
Thanks man......finally understood why we need convolution theorem
I can surely say that i am continuously happy to see you explaining this ideas. Thanks
So who is convolution? I still don’t get it.
Oh, it’s just the integral of f(y) g(x-y) dy, a neat way of multiplying functions
Dr Peyam
Lol I know. Notice I asked “who”. Since I remembered my still asked me who is convolution before. Because it’s taught after Laplace.
Self-similarity..... see my other postings in the comments, please!!
I just did! Thank you! That is so cool!
I wasn't really understanding convolution...just had a broad idea of it.... this video made my mind click😎🔥.. insane stuff
Thanks for the explanation! The convolution is quite intuitive to me now
Wow!, Thank You Very Much Sir....This Is A Very Nice Point Of View!!!
Damn, this is amazing, brother. Though I'll need to watch this video like 2 or 3 more times to connect the dots.
Keep up with the good work! Really, you are one of the most interesting and useful TH-cam channels I've been subscribed to
really helpful and save my life!!❤❤❤thank u 😊😊😊❤❤❤
You’re welcome 😊
Okay, I really am seeing what you did there, but I feel like what makes this really suggestive to me is looking at the each power of x as a basis function. Wooooow, this is so much more abstracted and interesting compared to the way I usually look at it as a moving inner product
You are so good!!!
You had me at "thanks for watching" 😍🤗
Great video!!! It really helps to make the intuition of convolution!
Great intuition. Please do more such videos.
Thank you so much for the baby step explanation ! It became completly intuitive thanks to the way you ve presented. It ! We want more of awsome content
Thanks for good video - from Korea
Sir, with your explaination, what was an esotheric formula, now has some real figure. Thank you very much!
Awesome stuff - thank you for the clear explanation
That's cool and gives a quite natural vector product for vectors in R^n:
(u*v)_i=Sum(0
Coool!!!
More interesting dear Professor
Thank you for this video, very helpful
You are welcome!
Very interesting. I wonder if we can draw other intuitions from polynomial functions and transfer them to general analytical functions. Anyway analytical function can be aproxymated locally by Taylor series. But in this case analogy seems to work not only locally but also in whole range.
:)
My mind went boom after this
Very neat explanation, thanks! 🤗
Dirty puns aside, really nice analogy. Never thought of it that way. As always, brilliant work.
I feel that this definitely helped me. Not really sure why you began discussing the continuous convolution though. The whole polynomial discussion is perfectly applicable in the context of discrete convolution. Anyway, for whatever reason, motivating with polynomial multiplication somehow did it for me. Thanks!
I'm also finding it helpful in higher dimensions to think in terms of multiplying polynomials (the number of variables = the number of dimensions): To find the coefficient for x_1^n_1 * x_2^n_2, you multiply coefficients of the input polynomials where the exponents add up to n_1 and n_2.
This kind of explains why you need to flip the "kernel" (in all dimensions) when you think of convolution as a "sliding dot product": when you flip the kernel, the coefficients that you need to multiply "pair up" (such that the exponents add up to n_i).
Also, I really like your sanity check: the two arguments MUST sum to x! Sounds gimmicky, but I'm pretty sure that will help me to remember.
thank you so much best one found
very helpful sir! thank you very much!
This was very useful, thank you so much
Subscribed! It was wonderful!
Thanks a lot doctor payam ! Convolution is really confusion topic for me ...
I would like to ask that,
is convolution useful for mathematicians. ..?
It is part of digital signal processing as per my information
So many applications! To get the distribution of the random variable X+Y, to solve Poisson’s equation, etc.
@@drpeyam Here's "the" classic video on convolution from the engineering school perspective: th-cam.com/video/_vyke3vF4Nk/w-d-xo.html
you will have to forgive the "cool 70s disco" look of the professor, it was indeed the 70s, so......he looks the part, (but, Prof Oppenheim is/was the guru on signals and systems theory") This is immensely useful math. immensely.
@@klam77 I agree with you. The "real" intuition of convolution comes from Signals and systems, the discrete case.
@@sandorszabo2470 Hello. But prof Peyam is nearly the same: when he talks of convolution in terms of multiplying two polynomials, prof oppenheim talks about "linear time invariant" systems which produce polynomial sums as "outputs" of multiple inputs in the LTI context! Almost similar! But yes, the original intuition was from the Engg department side, historically.
Just beautiful! thanks a lot!
very enjoyable! good stuff!
7:13 what do you mean with *f* and *g* as *"continuous polynomials"* ?
Think of a polynomial as an expression of the form sum a_n y^n and what I mean is an expression of the form sum a_x y^x where x ranges over the reals
Brother I know mechanical engineers could find resonance but when I had a deep thought on this resonance Is an slow accumulation of energy which is accumulated very high in small installments when the frequencies match if you strike a turning fork of 50 hz you get the same frequency of vibration on another tuning fork so they both vibrate if you strike it harder the amplitude changes hence loudness is a human factor the frequency is the same the languages that human speak through out the world the sound only resonate your ear drum for few seconds my question is that the harmonics is the fundamental frequency and overtones are the frequency that follow it take a word in any language you spell it according to convolution the thing scales and ques and stack the signal so convolution can be used to model resonance so when your ear drum vibrates it vibrates so the electrical signals are carried to brain like tuning fork ear drums vibrate within the audible spectrum 20 hz to 20000 hz hence resonance is caused by the word we speak and within the audible range the ear drums vibrate and we make sense of words I have seen in one videos on TH-cam that due to harmonics in any sound causes resonance which could be modelled by convolution recalling the resonance its destructive because slow and steady accumulation of sound on the mass causes high stress and high energy to build inside and stress increase and the system fractures or collapses but our ear drum hearing the sound from human languages try to vibrate but why our ear drum when subjected to continuous exposure of sound does not fracture or rupture like a wine glass iam not telling about high loud sound higher than 80 db but a audible range sound within the frequency of 20 hz to 20000 hz under continuous exposure why it's not damaging it again not failure by high energy but low one in synchronisation on air . But I tried it in my students when I told them to be quite in class they did not listen to me so I took my phone and set an frequency 14000 hz and they told it was irritating the idea of resonance is "small effort but large destruction " just like Tacoma bridge where the wind just slowly accumulated energy on the bridge and it collapsed it so my conclusion is if an audible frequency at continuous exposure to an human ear can it cause bleeding again "small effort but great destruction" sorry for the long story I you are able to reach hear you must be as curious as me so still not finished the ear drum is shook by harmonics in the sound we make by the words( or )overtones in the sound we make by the words I know harmonics is the fundamental frequency and overtones are following it which under slow and steady accumulation of sound energy resonates and could damge the ear drums again "small effort but big destruction" not to mention we assume the person is in coma or brain dead hence when the sound irritates him he or she could not make a move so my question is so simple normally human ear responds to harmonics or overtones according to convolution which could be a disaster but with minimal effort 🙏🙏🙏🙏 at here I could be wrong because harmonics can also be used to construct sound so can it be destructive or the overtones which are the trouble makers and which one according to this gives a response curve when two signals convolved by harmonics or overtones which is destructive but with minimal effort and convolution happens when ear drums oscillate is by harmonics or the overtones or also the trouble makers there
Convolution is integration of *f(τ).g(t-τ) dτ*
*(τ , τ+dτ)* is an infinitesimally small time period for which we are assuming the values of *f(τ)* and *g(t-τ)* remain constant.
*f(τ)* is the evolution of f(t) until the time instant *τ* .
*g(t-τ)* is the version of g(t) which came into existence at the time instant *τ* .
I understand that convolution is analogous to the multiplication of two polynomials. The intuition here is to express any signal f in terms of its impulses, just like coefficients of a polynomial. It makes sense, thanks. But I still do not understand why we convolute a signal when it is filtered. We may multiply the signal with the filter point-wise.
Awesome !!!!
Thanks for video De Peyam! Can you show Fourier and Laplace transform of convolution?
When I transform a function f(x )into an endless series, and also the function g(x)
Then I create a convolution with these two powerseries in your discrete way, with Sigma and Indices.
Is the resulting series the same as I transform the continous version (f*g)(x) into a series?
That was my realization from the video too. Now that I think about it, I think that's the result from multiplication of taylor series both fixed about a point a.
Hey Dr. Peyam. I had this issue in undergrad too! Thank you for the video. Out of curiosity, for convolutional neural networks, whenever they talk about the "window" in convolving images, would the "window" be analogous to getting the coefficient of a particular degree on this example?
love you sir
Remark 1: This one is pretty nice: th-cam.com/video/QmcoPYUfbJ8/w-d-xo.html
["What is convolution? This is the easiest way to understand" by Discretised]
It is in terms of integration of processes with fading intensity, but it is amenable for economic interpretation as well.
Remark 2: This multiplication by gathering indices that sum up to a constant is crucial for the Cauchy product of two infinite series instead of polynomials (Mertens theorem).
Remark 3: This convolution is with respect to time. In image manipulation the convolution is with respect to space (a kind of weighted averaging over pixels). That "spatial convolution" in the continuous case leads to an integral transform. One of the functions under convolution is then called a kernel. Just loose thoughts.
Is it called the convolution because it is convoluted?
Hahaha, probably! But I’m thinking more of “interlacing” values of f and g
Convolution is a term that really might better be described as “self-similarity”. It even has application to music theory! (THERE is the Group Theory connection!!! And even Biology!!!)
Hey Dr Peyam, can we perhaps see a more rigorous definition of what you mean by “continuous polynomials”, how functions can be described in terms of them, and how that leads to the convolution?
I would also love to see how this connects to the view of convolution in terms of linear functionals, as Physics Videos By Eugene made an extensive video on that which at least I didn’t really understand...
Anyhow, thanks a lot for this!
There is no rigorous definition of continuous polynomials, they don’t exist
@@drpeyam Couldn't we define them as an integral average?
like
the definite integral from zero to n of a(t)*x^t dt, all of that divided by n to "cancel out" the "dt" part, if we look at it from a perspective of dimensional analysis like done in physics
My profs at The Cooper Union, 1967-1971 likes to say the variable integrated over is “integrated out”..... which I hold is not accurate, as it is only in appearance, vanished..... the functions evaluated at each point of the “integrated out” variable contribute to the sum, as well as the end points. As the variable EXPLICITLY vanishes, it “goes away. By the way, Dr. Tabrizian, what is “f hat” you refer to in the Fourier transform description of the convolution? Please explain?
Fourier transform
Dr Peyam thanks!
Nice bro
Thank you very much.
What is the intuition behind the differential dy appearing as we transition to the continuous case?
It tells to integrate wrt y and keep x fixed (the resulting function is of x variable). Integration wrt y is a continuous analog of summation over the index (also termed y at the end of the presentation, to highlight the jump from a discrete to the continuous case).
so what is the result of the convolution of those two polinomials?
It made sense to me in some way...I still want to know the advantages of 'reflecting' and 'shifting' a function and then multiplying that with another function. If we do not 'reflect' then what ? Shifting I can understand..we have to keep moving window everywhere..
2:35 You should handle less coefficients and more coffeeicents ^^
Now I'm really curious if this interpretation could be used to give a more intuitive interpretation of volterra series analysis. Which is my favorite analysis technique that I learned in electrical engineering
This is math, simple and everything has to make sense.
My professor 7 years ago showed this by sliding a triangle into a rectangle until everything became convoluted
amazing
Wikipedia says the symbol from your thumbnail means something called ,,cross corelation'' and it's simmilar to convolution. I hope somewhere in future you will make a video about that.
Thank you
Does this mean if you convolute a function with 1 you get a taylor series?
That means you could get the taylor series of sin^2x, that would be useful in solving diff equations by solving for a taylor series. You could also continue value for sinx in the infinities.
Oh so i finally understood. F and g aren't time functions. They are the formula for the element of the taylor series. Sinx isn't f. F is i^(k-1)*(1/k!)* ( k mod 2)
I could never find an explanation of why (graphically) you have to reflect one function, multiply both and integrate. I see its too keep the indices to always sum the same?
brilliant
Thanks :)
do alternate notations for convolution exist. I hate that notation for convolution since i love using * to mean multiplicaiton and do so quite frequently.
I love *
I like your videos, but the whiteboard writing is not clear. It would be worthwhile to fix that because the content and presentation is good!
Actually you showed us the similarity between the two formulas , however I didn't understand convolution from that similarity 😥
Was watching this at 1.5x and laughed out loud
It was amazing but at the moment you replaced coefficient with the function itself, I didn't understand actually how you did this. Is there anyone who can make it clear for me? Thanks.
Convolution in one Word ???? Please answer!!!
a_k*b_x-k looked like a Cauchy Product. Is this a coincidence?
Sir I cant seem to practice this subject the right way......I'm worried the question might get twisted in the exam and my brain will freeze
It's hard to believe that you can take out time from your schedule to answer TH-cam comments
Thank you! :)
Why convolution is used?
How can i write a continuous polinomium:?
With convolution :)
Dr Peyam what a convoluted reply!!! :)
I think analogous to normal polynomial :
P(x) = integral(a(t)*x^t)dt
Patryk49 what is a normal polynomial? Is there a correspondence to a NormalSubgroup?
Here’s one answer: mathworld.wolfram.com/NormalPolynomial.html
Huh? 🤔 interesting.
convolution is just multiplication in the group algebra!
Just what I was thinking! I only recently started reading up on group modules and thus getting my jaw slowly pulled down whilst watching this
forgetful functor please explain or at least partially illuminate the Group Theory connection?
@@dougr.2398 If I may, this may help (straight to the examples in the wikipedia article about group rings): en.wikipedia.org/wiki/Group_ring#Examples
This is good. But it feels like a start, and the goal of a convolution is not explained. why do so, why use polynomial coefficients?
Wow
Vous avez un bon accent Français!
Merci!
If you replace x with e^(I th) you see convolution theorem.
You were right to both hesitate and then ignore the possibility that you had misspelled “coefficients”. English is difficult because it is FULL of irregularities.... this is one instance of a violation of the rhyme “I before E (edited 12-12-2023) except after C or when sounding like “Eh” (“long” A) as in Neighbor and Weigh”. Had you bothered to worry about that during the lecture, it would have impeded progress and the continuity (smile) of the discussion.
I do not take drugs. Never have. But now I feel like I’m on drugs. What’s the point of all this??