I'm currently in a graduate numerical methods course for mechanical engineers and I must say, as someone who doesn't have a math background, the visual analogy was very helpful! Thank you, you've brought me one step closer to grasping this stuff!
Hello. First thanks for the vivid video and explanation! I just have a doubt about the formula of slope shown at 2:19 where slope = delta_x / delta_y. I think it should be slope = delta_y / delta_x
thank you for that :) i would like to know why (0,0) point is the minimum value ? what if we choose a values that goes in -x and -y , is that mean the (0,0) is not always the minimum value ?
That's correct, the value shown in the video is a local minimum, and the function is unbounded for large negative x and y values. It's a good example of why it's important to understand the shape of the function you are optimizing, and one of the potential downfalls of gradient based solvers. This example would technically be more correct if posed as a constrained optimization problem to limit the search to just the area shown.
Thank you so much for such a helpful video! May I ask what software setup you used to create illustrations and images on the screen and to edit the video?
Thanks Sungwon, I'm glad you enjoyed it. I make these videos in PowerPoint with Matlab and Python graphs included when needed. The new morph transition in PowerPoint is especially helpful. I record the presentation and voice-over in OBS Studio, and do some touch up in Adobe Premiere before publishing.
I'm currently in a graduate numerical methods course for mechanical engineers and I must say, as someone who doesn't have a math background, the visual analogy was very helpful! Thank you, you've brought me one step closer to grasping this stuff!
simple and the best to grab the concept. thanks for this video
many thanks! please upload more videos! there are very helpful, especially for students!
Your videos are excellent! 10 / 10 !
Hello. First thanks for the vivid video and explanation! I just have a doubt about the formula of slope shown at 2:19 where slope = delta_x / delta_y. I think it should be slope = delta_y / delta_x
Glad you enjoyed it Kevin. Yes, I mistyped the slope equation. I've noted the correction in the video description.
it helped me a lot! thank you for your video
Thank you!! Very useful primer
Thank you!
that is very helpful , thankyou
Nice 👍🏽
Nice video AlphaOpt. I would like to help to point out that the slope equation in 2:20 is mistyped. I hope it helps. Thanks.
Nice catch JCOp, I've noted the correction in the video description.
thank you for that :)
i would like to know why (0,0) point is the minimum value ? what if we choose a values that goes in -x and -y , is that mean the (0,0) is not always the minimum value ?
That's correct, the value shown in the video is a local minimum, and the function is unbounded for large negative x and y values. It's a good example of why it's important to understand the shape of the function you are optimizing, and one of the potential downfalls of gradient based solvers. This example would technically be more correct if posed as a constrained optimization problem to limit the search to just the area shown.
Hello sir ,isnt the slope =dy/dx
rise over sun?
Thank you so much for such a helpful video! May I ask what software setup you used to create illustrations and images on the screen and to edit the video?
Thanks Sungwon, I'm glad you enjoyed it. I make these videos in PowerPoint with Matlab and Python graphs included when needed. The new morph transition in PowerPoint is especially helpful. I record the presentation and voice-over in OBS Studio, and do some touch up in Adobe Premiere before publishing.
Thank you
thanks a lot
what's the meaning of solver here?
sorry, because my first language is not english , so i can't understand...
Hi Seungjun, in this context a solver is a computer program that finds the solution to an optimization problem.
you're so amazing
thank you please help me
Jesus Christ is Lord. God of the living.