Vidit Nanda Profile Banner
Vidit Nanda Profile
Vidit Nanda

@viditnanda

6,346
Followers
812
Following
452
Media
1,791
Statuses

prof @oxunimaths & fellow @pembrokeoxford .

Oxford, England
Joined July 2015
Don't wanna be here? Send us removal request.
Pinned Tweet
@viditnanda
Vidit Nanda
4 months
I've been asked to serve as a founding (co)editor-in-chief of a new mathematics journal. There is some freedom in setting journal policy, etc. What are some things that you want to see when it comes to dealing with journals? What do you hate about the process? Share widely!
14
9
47
@viditnanda
Vidit Nanda
3 years
If this tweet gets > 500 ♥️s, I will compile the entire set of Computational Algebraic Topology lecture notes (8 Chapters, 106 pages) into a single hyperref-ed PDF and share the link here. Pass it along.
13
66
1K
@viditnanda
Vidit Nanda
3 years
I am teaching an 8-week course at Oxford on the fundamentals of topological data analysis (including persistent homology, cellular sheaves, & discrete Morse theory). All Notes: All Videos: Course thread! (0/8)
Tweet media one
16
187
941
@viditnanda
Vidit Nanda
15 days
"This talk is applied applied topology --- I will be applying applied topology to topology." - Shmuel Weinberger, today.
Tweet media one
15
54
895
@viditnanda
Vidit Nanda
7 months
Tweet media one
14
71
740
@viditnanda
Vidit Nanda
3 years
In lieu of my soundcloud, here's the promised pdf containing all the lecture notes for the Computational Algebraic Topology course: Curious about topological data analysis, persistent homology, cellular sheaves or discrete Morse theory? Check them out!
Tweet media one
@viditnanda
Vidit Nanda
3 years
If this tweet gets > 500 ♥️s, I will compile the entire set of Computational Algebraic Topology lecture notes (8 Chapters, 106 pages) into a single hyperref-ed PDF and share the link here. Pass it along.
13
66
1K
16
157
731
@viditnanda
Vidit Nanda
5 months
Inside you are 2' wolves
Tweet media one
10
31
568
@viditnanda
Vidit Nanda
1 year
Topological Data Analysis as you've never seen it before: @robertghrist is bringing my lecture notes () to life in a new animated video series ()! Here are the videos which have been released so far... enjoy! And tell your friends.
6
116
496
@viditnanda
Vidit Nanda
2 years
I've made an offer to buy the Annals of Mathematics for $44 billion. Your papers will be published if and only if you pay 8 bucks a month.
40
38
478
@viditnanda
Vidit Nanda
2 years
You: just starting to feel safe in the privacy of your own home… Mathematics:
Tweet media one
1
58
457
@viditnanda
Vidit Nanda
5 years
New hobby: tricking data scientists into learning representation theory by calling it grp2vec
9
59
424
@viditnanda
Vidit Nanda
7 months
Last week, I co-organised a workshop at the lovely @NewtonInstitute in Cambridge. We tried to do a few things differently; some of these worked well, others didn't. Here are two principles that worked well. 1/6
Tweet media one
10
34
409
@viditnanda
Vidit Nanda
1 year
Say what now? 🍻 @robertghrist
Tweet media one
51
5
368
@viditnanda
Vidit Nanda
4 years
Like many of you, I've recently had to make my peace with asynchronous lectures. This term, I'm teaching Computational Algebraic Topology @OxUniMaths --- this course covers the foundations of topological data analysis and is meant for senior undergraduates. (1/4)
Tweet media one
8
63
366
@viditnanda
Vidit Nanda
4 years
Normalize "Conflicts of Interest" statements in Mathematics Papers :p
Tweet media one
3
64
364
@viditnanda
Vidit Nanda
8 days
̷A̷s̷s̷o̷c̷i̷a̷t̷e̷ Professor of Mathematics
48
3
462
@viditnanda
Vidit Nanda
2 years
Annual reminder that I am teaching this topological data analysis course again this term @OxUniMaths , and *all of the materials are already online* so you can follow along at your pace :)
@viditnanda
Vidit Nanda
3 years
I am teaching an 8-week course at Oxford on the fundamentals of topological data analysis (including persistent homology, cellular sheaves, & discrete Morse theory). All Notes: All Videos: Course thread! (0/8)
Tweet media one
16
187
941
8
48
349
@viditnanda
Vidit Nanda
3 years
Just found out that the entire plot of The Matrix 5: Rotations has been leaked online --- [ cos t -sin t ] [ sin t cos t ]
12
29
339
@viditnanda
Vidit Nanda
2 years
Tweet media one
9
4
304
@viditnanda
Vidit Nanda
3 years
Outstanding bit of notation that I've just seen in a graduate student's thesis writeup: B_r(x) for open ball of radius r around x, and B_r[x] for closed ball. Don't know if he got it from somewhere, but I'm kicking myself for not thinking of this for my own papers...
16
19
284
@viditnanda
Vidit Nanda
4 years
The best harmonic sequence on the planet: a thread (1/n)
6
41
277
@viditnanda
Vidit Nanda
4 years
I can't remember where I read this: "doing mathematics is like climbing a cliff that is vertical in front of you and horizontal behind you" 🙄
@csaez_math
Carles Sáez
4 years
The two stages of math problem solving, or why math is so depressing: 1. This problem is so hard, I don't even know where to start. I must be stupid. 2. The solution was so obvious I can't believe I spent so much time without seeing it. I must be stupid.
33
259
1K
5
30
270
@viditnanda
Vidit Nanda
1 year
Tweet media one
4
32
272
@viditnanda
Vidit Nanda
11 months
My reaction whenever a paper introduces some m x n real matrix as A ∈ |R^{m x n}
25
8
266
@viditnanda
Vidit Nanda
4 years
Happy to report that my lecture on entropy is more disorganized than it was last year.
3
15
264
@viditnanda
Vidit Nanda
3 years
I was getting worried about the new dirac-delta variant, but thankfully you can only catch it from yourself.
14
23
262
@viditnanda
Vidit Nanda
2 years
Many mathematicians, whether they confess it publicly or not, have a "Tombstone Theorem" --- the result you'd want on your tombstone if you were to pass away today. Here's mine. (0/8)
Tweet media one
8
31
261
@viditnanda
Vidit Nanda
5 years
Kashiwara-Schapira: making grown men scream into their pillows in despair since 1990.
Tweet media one
17
52
246
@viditnanda
Vidit Nanda
3 years
I'm fascinated by this amazing phase transition that occurs around year 2 or 3 of graduate school. All meetings before this magic moment involve me explaining things to my grad students, and all meetings after that involve them explaining things to me.
4
16
247
@viditnanda
Vidit Nanda
3 years
Here are twelve pages on #PersistentHomology , which is the focus of Week 6 in the Computational Algebraic Topology course. I'm giving all my secrets away :) Course webpage if you want to see older notes to chase backreferences!
Tweet media one
8
57
247
@viditnanda
Vidit Nanda
1 year
Also: 1) You don't like the explanation you read 2) With much effort you figure out a clear explanation 3) It is basically identical to the original explanation you disliked, but somehow it makes sense to you now
@Singularitarian
Daniel O'Connor
1 year
How learning math works: 1) You don’t like the explanation you read. 2) With much effort you figure out a clear explanation. 3) The explanation you came up with is obvious to experts but nobody ever mentions it.
40
190
2K
5
27
240
@viditnanda
Vidit Nanda
3 years
There are many good answers, but this one satisfies a universal property.
Tweet media one
@Variety
Variety
3 years
‘Unfilmable’ Books That Hollywood Should Try to Adapt After ‘Dune’
235
58
510
7
32
242
@viditnanda
Vidit Nanda
4 years
ProTip: if you want to learn everything there is to know about derived categories, just draw this sigil on your mirror at midnight and chant “RHom Verdier” three times.
Tweet media one
5
35
230
@viditnanda
Vidit Nanda
5 years
The first lecture of computational algebraic topology is almost ready. Needs more cowbell.
Tweet media one
5
31
216
@viditnanda
Vidit Nanda
1 year
It's finally online!
Tweet media one
@viditnanda
Vidit Nanda
2 years
I usually avoid writing papers that either (1) have lots of co-authors, or (2) contain no genuinely new ideas, or (3) require me to read *many* of existing papers. This new paper violates all three rules... 1/7
Tweet media one
4
28
144
2
38
206
@viditnanda
Vidit Nanda
2 years
Many years ago, I saw a much-beloved Mathoverflow question by @wtgowers on "Common False Beliefs in Mathematics": Last week I realized that I held a false belief. I don't know if it's common, but it is downright INFURIATING. (1/7)
Tweet media one
2
25
204
@viditnanda
Vidit Nanda
4 years
Introducing 🔥GeoTorch🔥, my student Mario L Casado's cool new library for optimization on interesting Riemannian manifolds. It handles Stiefel varieties, Grassmannians, and Lie groups! If you're thirsty for deep learning on SO(n), check it out :)
2
55
198
@viditnanda
Vidit Nanda
3 years
Minor update: (👶,👩,👨)
@viditnanda
Vidit Nanda
4 years
16
3
143
23
0
191
@viditnanda
Vidit Nanda
2 years
Symplectic geometers Lin-Manuel Miranda 🤝 Hamiltonian flow
3
11
182
@viditnanda
Vidit Nanda
3 years
It's been a while, mathjobs dot org.
Tweet media one
6
12
171
@viditnanda
Vidit Nanda
2 years
1/ Here are two stories about my graduate school experience at Rutgers, both about a man who has played a much larger role in my career than he realised.
3
20
176
@viditnanda
Vidit Nanda
2 years
I've seen so much terrible academic advice on this website over the last few days (thanks to this wretched default "For you" tab) that I have decided to start offering my own. 1. Don't waste time studying the p-adics, their p-value is always larger than 0.05
7
14
176
@viditnanda
Vidit Nanda
5 years
(Almost) everything you wanted to know about persistent homology, but were afraid to ask:
Tweet media one
5
40
171
@viditnanda
Vidit Nanda
4 years
The notes for Week 3 of Computational Algebraic Topology contain one weird homology trick, recently discovered by a PhD student (tenured profs hate it!)
Tweet media one
7
22
170
@viditnanda
Vidit Nanda
3 years
Very cool new paper by Jabob Leygonie and Greg Henselman-Petrusek on the following inverse problem: given a barcode, find a filtered simplicial complex whose persistent homology realizes that barcode.
Tweet media one
7
33
166
@viditnanda
Vidit Nanda
2 years
cc: @wtgowers How it started How it's going
Tweet media one
Tweet media two
2
18
162
@viditnanda
Vidit Nanda
2 years
Analytic number theorists Lumberjacks 🤝 log log log
2
11
155
@viditnanda
Vidit Nanda
5 months
Just found this devastating critique of the last few seasons of Game of Thrones on a Japanese website.
Tweet media one
7
16
155
@viditnanda
Vidit Nanda
4 years
The course starts on Monday --- in Week 1 we will cover simplicial complexes, geometric realizations, simplicial maps, and how to associate filtrations to finite datasets. The journey begins here: Notes: Videos:
Tweet media one
6
30
148
@viditnanda
Vidit Nanda
4 years
Week 4 of Computational Algebraic Topology is here! These lecture notes serve as an appetizer for both category theory and homological algebra en route to describing powerful tools for working with simplicial homology!
Tweet media one
3
32
151
@viditnanda
Vidit Nanda
5 years
Lecture 3 of computational algebraic topology is up! Here we meet Euler characteristics, chain complexes and homology. If you make it to the end, there's ONE WEIRD PICTURE THAT EXPLAINS EVERYTHING. OTHER IMAGES HATE IT.
Tweet media one
5
29
146
@viditnanda
Vidit Nanda
3 years
Homologists of the world, unite! You have nothing to lose but your chains. The final week of Computational Algebraic Topology involves discrete Morse theory, a wonderful tool for homology-preserving reduction of large chain complexes. Notes:
Tweet media one
4
20
145
@viditnanda
Vidit Nanda
4 years
16
3
143
@viditnanda
Vidit Nanda
2 years
I usually avoid writing papers that either (1) have lots of co-authors, or (2) contain no genuinely new ideas, or (3) require me to read *many* of existing papers. This new paper violates all three rules... 1/7
Tweet media one
@majirouses
Maria Jose Jimenez
2 years
After a great collaboration to join efforts ... here it is: our Xmas present for TDA practitioners🥳 It was a big pleassure to work with Dashti Ali, @Aras121 , @viditnanda , @eduph_tw and Manu Soriano-Trigueros
2
10
58
4
28
144
@viditnanda
Vidit Nanda
2 years
I definitely took a wrong turn several pages ago ☠️
Tweet media one
8
5
136
@viditnanda
Vidit Nanda
1 year
Tis better to have Liebed and Lost than never to have Liebed at all.
Tweet media one
4
14
136
@viditnanda
Vidit Nanda
4 years
Homology has an evil twin (with superpowers). In the fifth week of Computational Algebraic Topology, we finally run into cohomology groups:
Tweet media one
2
22
134
@viditnanda
Vidit Nanda
5 years
"A topologist is someone who understands that the food you eat is never really inside you." 👆a far better descriptor than coffee cup = donut.
3
38
129
@viditnanda
Vidit Nanda
4 years
I have to read like 300 postdoc applications this morning. Just wanted convey a big thank-you to the wonderful person who opened their research statement with "My research interests fall into a few main categories, most of which are Abelian."
2
4
131
@viditnanda
Vidit Nanda
4 months
Baby #2 is due any day now, and I am too wired to do the intelligent thing (i.e., sleep). So please join me, the wonderful @osumray , and @haharrington on a journey through some fun machine learning & representation theory. 1/11
Tweet media one
3
26
121
@viditnanda
Vidit Nanda
3 years
It is an edict of the Archangel Gabriel that directed graphs shall be called "quivers" before one can do any representation theory on them. Here's a short thread to celebrate a new preprint () with @annaseigal and @haharrington . (1/7)
Tweet media one
4
34
119
@viditnanda
Vidit Nanda
4 years
This number seems to me a rather dull one, and I hope it is not an unfavorable omen ; )
Tweet media one
5
4
115
@viditnanda
Vidit Nanda
3 years
If you prefer the movie to the book, here are four video lectures on Persistent Homology :)
@viditnanda
Vidit Nanda
3 years
Here are twelve pages on #PersistentHomology , which is the focus of Week 6 in the Computational Algebraic Topology course. I'm giving all my secrets away :) Course webpage if you want to see older notes to chase backreferences!
Tweet media one
8
57
247
4
26
117
@viditnanda
Vidit Nanda
4 years
I should be proving theorems, but I can't stop thinking about a hypothetical restaurant which exclusively serves appetizers in large portions and is called "Hors d'oeuvres of Magnitude".
4
12
114
@viditnanda
Vidit Nanda
3 years
@littmath writing a new paper whose i-th word is different from the i-th word of Cantor's i-th paper.
0
11
115
@viditnanda
Vidit Nanda
4 years
Ever tried to sneak sheaves, derived categories and local cohomology into @PNASNews ? Not easy, but with help from outstanding co-authors it can be done. Thanks to @haharrington , Jared Tanner, and most of all, to Bernadette Stolz!
2
31
115
@viditnanda
Vidit Nanda
4 months
2 Minor 2 Update: (👶,👦, 👩‍🦰,👨)
@viditnanda
Vidit Nanda
3 years
Minor update: (👶,👩,👨)
23
0
191
18
1
115
@viditnanda
Vidit Nanda
1 year
Am I doing this right, @miniapeur ?
Tweet media one
2
5
106
@viditnanda
Vidit Nanda
3 years
I love all of you very much, so please don't beat me up for this.
Tweet media one
9
23
107
@viditnanda
Vidit Nanda
3 years
This is how I roll now 😎
Tweet media one
3
0
102
@viditnanda
Vidit Nanda
5 years
Overfull \hbox is my tragic flaw.
Tweet media one
2
14
100
@viditnanda
Vidit Nanda
2 years
Algebraists Statisticians 🤝 R-modules
7
4
99
@viditnanda
Vidit Nanda
3 years
I respect all my mathematics textbooks, especially Differential Forms in Algebraic Topology by Bott and Usted.
9
4
96
@viditnanda
Vidit Nanda
11 months
Special Announcement: the Bourbakistas and von Neumenn have logged on to tell us that ordinary multiplication of positive integers doesn't actually commute.
Tweet media one
10
6
92
@viditnanda
Vidit Nanda
3 months
One of the students whose final exams I am grading has left one of the calculations "as an exercise for the examiner".
3
2
95
@viditnanda
Vidit Nanda
7 months
The news: The Persistent Homologist:
Tweet media one
Tweet media two
0
10
97
@viditnanda
Vidit Nanda
3 years
Interviewing PhD candidates born after 1/1/2000 next week.
4
0
93
@viditnanda
Vidit Nanda
2 years
Recently had the amazing experience of being completely disarmed by an anonymous referee report. After this paragraph, even if the referee had recommended that I "replace all the words with new ones," I would have done so with a smile on my face.
Tweet media one
1
0
92
@viditnanda
Vidit Nanda
6 years
Man, the local cohomology store sure has some great products.
Tweet media one
5
22
88
@viditnanda
Vidit Nanda
1 year
Very cool, but they could have saved a lot of time and gpu costs by simulating the same black hole with microsoft paint:
Tweet media one
@NOIRLabAstro
NOIRLab
1 year
The iconic image of the supermassive black hole at the center of Messier 87, has received its first official makeover, thanks to a new machine-learning technique known as PRIMO! 🧵⬇️
Tweet media one
5
65
256
5
1
90
@viditnanda
Vidit Nanda
2 years
A new picture-filled thread for a new paper! This is joint work with Ulrike Tillmann and our awesome student Marc Fersztand. To my mind, it is the first point of contact between topological data analysis and geometric invariant theory. 1/12
Tweet media one
@mathRTb
arXiv math.RT Representation Theory
2 years
Marc Fersztand, Vidit Nanda, Ulrike Tillmann: Harder-Narasimhan Filtrations and Zigzag Persistence
1
1
8
2
16
90
@viditnanda
Vidit Nanda
1 year
Mathematicians on the academic job market this year: please reply with a description of (or link to) your work! If there are enough responses, I will leave this tweet pinned for the next few months and hopefully more people will get to hear about you. Good luck 👍
1
22
86
@viditnanda
Vidit Nanda
3 years
Omg the computational algebraic topology notes are on Reddit! Who did this??
2
5
88
@viditnanda
Vidit Nanda
2 years
I'm co-organising a 6 month program @NewtonInstitute on all things equivariant & geometric in early 2024! If you are interested in attending some or all of this — esp. Workshop 2 — please drop me a line, and tell all your friends :)
Tweet media one
8
21
86
@viditnanda
Vidit Nanda
4 years
[1/20] Complexes of groups are concrete objects living at the intersection of many abstract fields of mathematics: algebraic topology, algebraic+differential geometry, and even higher categories. But their most natural home is in geometric group theory...
4
17
83
@viditnanda
Vidit Nanda
3 years
I've never understood this weird stereotype of the British being a pessimistic people. You can't walk 15 minutes along any path in Oxford without stumbling across a bloody sundial. If that's not a sign of unrelenting optimism, I don't know what is.
Tweet media one
2
6
80
@viditnanda
Vidit Nanda
4 years
Homotopy theory has a formidable (and richly deserved!) reputation: too abstract, hard to compute... But large parts can be made concrete, and help simplify computation of weaker topological invariants (eg cohomology). Homotopy is the focus of Week 2 of the TDA course! (1/3)
Tweet media one
1
18
83
@viditnanda
Vidit Nanda
4 years
We really need to start using thenn for "then and only then"
4
6
79
@viditnanda
Vidit Nanda
5 years
Note to shelf: why call it a bunch of algebra textbooks when you could be calling it your artin stack?
@aacooper
Cooper
5 years
Tweet media one
Tweet media two
2
0
18
2
4
79
@viditnanda
Vidit Nanda
3 years
Week 7 of computational algebraic topology is all about sheaves! Check out the notes (if you dare) You will meet sheaf cohomology, etale spaces, sections... And I think the title picture is @robertghrist 's best one yet
Tweet media one
2
16
79
@viditnanda
Vidit Nanda
4 years
Care to participate in another experiment? I've never done a thread on twitter before, and I really want to *show* you one of my favorite mathematical objects. Sounds like a two-for-one; let's go! 1/13
2
10
76
@viditnanda
Vidit Nanda
1 month
@littmath "Jake, the senator has repeatedly mentioned that he wants the states to have more powers. I agree, but we need to go further. The best system of governance is one where each community locally decides what is best for it. In fact, let L be such a local system..."
0
4
77
@viditnanda
Vidit Nanda
3 years
In 3 days I will release an amazing new preprint on the topology of cryptocurrency protocols. Prepare yourselves for blockchain homotopy equivalence.
4
1
77
@viditnanda
Vidit Nanda
3 years
The new James Bond movie's looking pretty good.
Tweet media one
2
13
74
@viditnanda
Vidit Nanda
2 years
The best thing about microsoft teams is the amazing chat feature, which can be used to send zoom links.
3
0
72
@viditnanda
Vidit Nanda
3 years
A brief thread about this shiny new paper with colleague Harald Oberhauser and our wonderful PhD student Uzu Lim. Here we obtain precise bounds on the number of points required for estimating tangent spaces of smooth manifolds with high confidence. (1/5)
@mathSTb
arXiv math.ST Statistics Theory
3 years
Uzu Lim, Vidit Nanda, Harald Oberhauser: Tangent Space and Dimension Estimation with the Wasserstein Distance
1
2
15
3
10
69
@viditnanda
Vidit Nanda
2 years
One of the great things about living in the UK is the ability to see some spectacular views during the winter afternoons. Here is the stunning vista from my office at 3pm yesterday:
Tweet media one
0
4
72
@viditnanda
Vidit Nanda
2 years
Homotopy theorists Ghostbusters 🤝 Chasing spectra
4
6
70
@viditnanda
Vidit Nanda
3 years
Just received a referee request by a journal I've never heard of for a paper well outside my area of expertise; when I click on the link to decline, it says I must create an account, log in, and then decline. What circle of hell is this?
5
1
70
@viditnanda
Vidit Nanda
3 months
"Sheaves on manifolds" by Kashiwara and Schapira.
@bernardtjoy
Bernard T. Joy
3 months
If neither Dante nor Milton were available and you needed a literary guidebook for a trip through Hell which literary guidebook would you choose?
235
11
345
3
3
67
@viditnanda
Vidit Nanda
2 years
Tweet media one
2
7
69