Geometric deep learning: First steps
a-j.gitbook.io
Geometric deep learning: First steps
1–10 of 10 posts
Re: Geometric deep learning: First steps
#2I don't have a clue about geometry and maths. BUT you've made it clearer than my entire educational system. thank you.
Btw, It's fascinating how simmetry can be applied in such a practical way.
Re: Geometric deep learning: First steps
#3Thank you for contributing this course.
That said, its legibility would be helped by some proofreading. Please accept these suggestions for improvements:
Don't randomly format words in bold, such as the word "vaguely".
"greek" -> "Greek"
"russian" -> "Russian"
"an hungarian" -> "a Hungarian"
reference: https://www.berlitz.com/es-mx/blog/gentilicios-en-ingles
Same rule for toponyms:
"greece" -> "Greece"
Greek proper names have different forms in English than in Spanish:
"Euclide" -> "Euclid"
reference: https://en.wikipedia.org/wiki/Euclid
Join sentence fragments:
"packing of particles. It was" -> "packing of particles, it was"
Capitalization, inconsistent spelling:
"euclidean", "Euclidian" -> "Euclidean"
"non-euclidian" -> non-Euclidean
Incorrect word choice:
replace: "This fifth postulate posteriorly defined any attempt to try to prove it from other postulates of geometry."
with: "This fifth postulate subsequently defied any attempt to try to derive it from the other postulates of geometry."
"19 th" -> "19th"
Unnecessary capitalization:
"Alternative proposed Postulate" - > "Alternative proposed postulate"
Incorrect proposition:
"Academia rejection to the proposal" -> "Academic rejection of the proposal:"
Typographical error:
"universitary" -> "university"
"Different persons came to the same idea from differents backgrounds" -> "Different people arrived at the same idea from different backgrounds."
"Standar" -> "Standard"
Contrast with the Spanish word "estandar".
"erlangen programme" -> "Erlangen program"
"EXISTENTS" -> "existence"
"lipschitz continuous" -> "Lipschitz continuous"
"elemen" -> "element"
Re: Geometric deep learning: First steps
#4I was literally toying around with some of these ideas today. I had the thought that you could take some time series data, compute the velocities and positions over time, and use that to try and learn the metric tensors of it, assuming the time series data is following a geodesic on some manifold for the most part. And from there you could compute the symmetries and curvature of the system. I thought you could use that to try and define an energy for the system or some other conserved quantities, and then use that as an anomaly detector (change in the energy of the system means there was some outside interference).
Re: Geometric deep learning: First steps
#5Thank you for contributing this course. That said, its legibility would be helped by some proofreading. Please accept these suggestions for improvements: Don't randomly format words in bold, such as the word " vaguely ". "greek" -> "Greek" "russian" -> "Russian" "an hungarian" -> "a Hungarian" reference: https://www.berlitz.com/es-mx/blog/gentilicios-en-ingles Same rule for toponyms: "greece" -> "Greece" Greek proper…
I'm hugely grateful that you took a detailed and meticulous reading of the article to find out such a large list of mistakes. Being honest, I already thought that I should have revisited with caution the legibility and shape of my language there before any submission. But my desire to share this little advance won the battle, I apologize for such a shame.
I have updated the doc according to your suggestions, and including several others. I will indeed try to formalise the article early tomorrow.
Once again, I can't thank you enough for your detailed feedback.
Re: Geometric deep learning: First steps
#6Thank you for contributing this course. That said, its legibility would be helped by some proofreading. Please accept these suggestions for improvements: Don't randomly format words in bold, such as the word " vaguely ". "greek" -> "Greek" "russian" -> "Russian" "an hungarian" -> "a Hungarian" reference: https://www.berlitz.com/es-mx/blog/gentilicios-en-ingles Same rule for toponyms: "greece" -> "Greece" Greek proper…
I forgot to address the issue related to my indiscriminate usage of the bold. I have always been an advocate of the fast reading techniques but you are right telling me that it may be too much of it.Note: I thought that I also linked the github hosting this gitbook to further collaboration. I have also upgraded that issue.
Re: Geometric deep learning: First steps
#7I was literally toying around with some of these ideas today. I had the thought that you could take some time series data, compute the velocities and positions over time, and use that to try and learn the metric tensors of it, assuming the time series data is following a geodesic on some manifold for the most part. And from there you could compute the symmetries and curvature of the system. I thought you could use th…
Unbelievable. Your ""toy"" experimentation led you to such a intriganting and curious mechanism to time series forecasting. Regardless of what I tell you, I wouldn't be capable of expressing my interest in your statement. It might be exciting to move forward and dive deeper in your idea. I will definitely jump on it.
Re: Geometric deep learning: First steps
#8I was literally toying around with some of these ideas today. I had the thought that you could take some time series data, compute the velocities and positions over time, and use that to try and learn the metric tensors of it, assuming the time series data is following a geodesic on some manifold for the most part. And from there you could compute the symmetries and curvature of the system. I thought you could use th…
I was looking at a similar approach, my question is how do parameterize the manifold and how do you map points on the manifold if it's irregular -think filamentous, so that path trace can be done
And in general this is a question of interest. Are graphs really a good discrete approximation of a manifold?
Re: Geometric deep learning: First steps
#9I was literally toying around with some of these ideas today. I had the thought that you could take some time series data, compute the velocities and positions over time, and use that to try and learn the metric tensors of it, assuming the time series data is following a geodesic on some manifold for the most part. And from there you could compute the symmetries and curvature of the system. I thought you could use th…
Would you like to explore this and other ideas related to the field and contemplate some kind of potential collaboration?
Re: Geometric deep learning: First steps
#10I was literally toying around with some of these ideas today. I had the thought that you could take some time series data, compute the velocities and positions over time, and use that to try and learn the metric tensors of it, assuming the time series data is following a geodesic on some manifold for the most part. And from there you could compute the symmetries and curvature of the system. I thought you could use th…
I was looking at a similar approach, my question is how do parameterize the manifold and how do you map points on the manifold if it's irregular -think filamentous, so that path trace can be done And in general this is a question of interest. Are graphs really a good discrete approximation of a manifold?
I hope you find these two resources useful to address your question. I'm also pretty integrated about the possible answers, but I definitely bet on doing it using graphs. I'm preparing a detailed and concrete explanation on why I guess this conclusion but that one will need to wait for now!
For now as I have already told you, check out these ones.