MY brother, I don't know if you'll see my comment, but this video was beautifully done. The presentation, the visuals, absolutely fantastic. Topology has always been intriguing to me, and it's something I really want to pursue, hopefully at a Masters and Doctoral level. Thank you for taking the time and effort to share this gem with us! 10/10 and subscribed!
Hello noah, I am doing a project in TDA and would like to talk to you about it, your explanation is very clear and you could help a lot! Thanks so much!
Hey I don’t know if you still would like any help, but I actually have worked with TDA for a couple projects now and have a pretty good conceptual understanding of it, and actually worked on trying to predict crashes in stocks with TDA in the past. The plots in the top right towards the end of the video were pulled directly from a paper that I followed in implementing my methods! I didn’t end up having very good predictive accuracy, but I did this project quite early in my data scientist journey and believe I have learned a lot about the do’s and dont’s when trying to use methods from TDA. Just send me a dm if you’re interested :)
Hey David, here a few resources that helped me. To get an introductory sense of things, I think it’s useful to go through the Wikipedia page and this Medium post: towardsdatascience.com/topological-data-analysis-tda-b7f9b770c951. Here are a few cool papers that use TDA in interesting (and different) ways. This is just the tip of the iceberg in terms of applications. 1) Gidea, Katz. “Landscapes of Crashes” (2017) -> uses TDA, specifically the L-p norm of the persistence landscape, to find early-warning signals of financial crashes such as the 2008 Recession and the dot-com bubble (this is the paper I tangentially discuss in my video). 2) Goldfarb. “Hockey Analytics” (2014) -> finds correlation between team offensive performance and specific persistence bar-code characteristics, suggests how to use this info to make a better player trade 3) Deng, Duzhin (2022). “TDA Helps Fake News Detection” --> finds that a deep learning NLP model trained on a small dataset can be improved using TDA If you want to really dig into the mathematical background, I recommend Hatcher's "Algebraic Topology" textbook, or this rigorous review by Larry Wasserman: arxiv.org/pdf/1609.08227.pdf. Hope this helps!
MY brother, I don't know if you'll see my comment, but this video was beautifully done. The presentation, the visuals, absolutely fantastic. Topology has always been intriguing to me, and it's something I really want to pursue, hopefully at a Masters and Doctoral level. Thank you for taking the time and effort to share this gem with us! 10/10 and subscribed!
Beautifully done with strong visualization of the concepts
Absolutely amazing video! Subscribed.
Very concise, i like it.
But how exactly those persistent holes represents financial crashes?? (bro lit missed the best part)
Really good explanation!!
Hello noah, I am doing a project in TDA and would like to talk to you about it, your explanation is very clear and you could help a lot! Thanks so much!
Hey I don’t know if you still would like any help, but I actually have worked with TDA for a couple projects now and have a pretty good conceptual understanding of it, and actually worked on trying to predict crashes in stocks with TDA in the past. The plots in the top right towards the end of the video were pulled directly from a paper that I followed in implementing my methods! I didn’t end up having very good predictive accuracy, but I did this project quite early in my data scientist journey and believe I have learned a lot about the do’s and dont’s when trying to use methods from TDA. Just send me a dm if you’re interested :)
Hi Noach, feel free to reach out at njbergam@gmail.com, I would love to hear more about what you're doing!
Cool.
Thanks for the video.
Hi Noah, great explanation! Any resources that are out there to get familiar to tda?
Hey David, here a few resources that helped me.
To get an introductory sense of things, I think it’s useful to go through the Wikipedia page and this Medium post: towardsdatascience.com/topological-data-analysis-tda-b7f9b770c951.
Here are a few cool papers that use TDA in interesting (and different) ways. This is just the tip of the iceberg in terms of applications.
1) Gidea, Katz. “Landscapes of Crashes” (2017) -> uses TDA, specifically the L-p norm of the persistence landscape, to find early-warning signals of financial crashes such as the 2008 Recession and the dot-com bubble (this is the paper I tangentially discuss in my video).
2) Goldfarb. “Hockey Analytics” (2014) -> finds correlation between team offensive performance and specific persistence bar-code characteristics, suggests how to use this info to make a better player trade
3) Deng, Duzhin (2022). “TDA Helps Fake News Detection” --> finds that a deep learning NLP model trained on a small dataset can be improved using TDA
If you want to really dig into the mathematical background, I recommend Hatcher's "Algebraic Topology" textbook, or this rigorous review by Larry Wasserman: arxiv.org/pdf/1609.08227.pdf.
Hope this helps!
Too bad it only works retroactively. You'd get a noble prize in economics if this worked