Deep Patel
@nolongerconvex.bsky.social
CS PhD at UW-Madison | Interests: Learning Theory, Trustworthy ML | Previously: IISc, Bangalore
Website: https://dbp1994.github.io
Website: https://dbp1994.github.io
Reposted by Deep Patel
A great slide from a great talk by @pseudomanifold.topology.rocks at @logml.bsky.social
July 9, 2025 at 11:05 AM
A great slide from a great talk by @pseudomanifold.topology.rocks at @logml.bsky.social
Reposted by Deep Patel
System is so broken:
- researchers write papers no one reads
- reviewers don't have time to review, shamed to coauthors, use LLMs instead of reading
- authors try to fool said LLMs with prompt injection
- evaling researchers based on # of papers (no time to read)
Dystopic.
- researchers write papers no one reads
- reviewers don't have time to review, shamed to coauthors, use LLMs instead of reading
- authors try to fool said LLMs with prompt injection
- evaling researchers based on # of papers (no time to read)
Dystopic.
July 7, 2025 at 4:15 PM
System is so broken:
- researchers write papers no one reads
- reviewers don't have time to review, shamed to coauthors, use LLMs instead of reading
- authors try to fool said LLMs with prompt injection
- evaling researchers based on # of papers (no time to read)
Dystopic.
- researchers write papers no one reads
- reviewers don't have time to review, shamed to coauthors, use LLMs instead of reading
- authors try to fool said LLMs with prompt injection
- evaling researchers based on # of papers (no time to read)
Dystopic.
Reposted by Deep Patel
This is a very good paper on how methodologies from comparative cognition can help improve AI evaluation:
openreview.net/pdf?id=gCPJF...
Other papers along the same lines:
⬇️
openreview.net/pdf?id=gCPJF...
Other papers along the same lines:
⬇️
openreview.net
June 20, 2025 at 7:58 PM
This is a very good paper on how methodologies from comparative cognition can help improve AI evaluation:
openreview.net/pdf?id=gCPJF...
Other papers along the same lines:
⬇️
openreview.net/pdf?id=gCPJF...
Other papers along the same lines:
⬇️
Reposted by Deep Patel
Fascinating lecture by Rudin—“Set Theory: An Offspring of Analysis”
www.youtube.com/watch?v=hBcW...
In brief, it describes how studying the wave equation led to trigonometric series, which eventually led to Cantor’s introduction of limit points and the notion of countable sets.
www.youtube.com/watch?v=hBcW...
In brief, it describes how studying the wave equation led to trigonometric series, which eventually led to Cantor’s introduction of limit points and the notion of countable sets.
Walter B. Rudin: "Set Theory: An Offspring of Analysis"
YouTube video by UW-Milwaukee Department of Mathematical Sciences
www.youtube.com
June 4, 2025 at 4:56 PM
Fascinating lecture by Rudin—“Set Theory: An Offspring of Analysis”
www.youtube.com/watch?v=hBcW...
In brief, it describes how studying the wave equation led to trigonometric series, which eventually led to Cantor’s introduction of limit points and the notion of countable sets.
www.youtube.com/watch?v=hBcW...
In brief, it describes how studying the wave equation led to trigonometric series, which eventually led to Cantor’s introduction of limit points and the notion of countable sets.
Reposted by Deep Patel
Nothing says upcoming NeurIPS deadline like Overleaf being down.
May 14, 2025 at 6:23 AM
Nothing says upcoming NeurIPS deadline like Overleaf being down.
Reposted by Deep Patel
Reposted by Deep Patel
Are you a mathematics teacher? Would you like a taste of 21st century mathematics? See the link below to join a program for teachers run by Justin Lanier @jlanier.bsky.social —stay for as long as you like, explore with friends, & get your own insights into accessible, deep questions in math
“21st century mathematics”: a free month-long online enrichment program for math teachers.
Info + register: bit.ly/21math2025
Video part 1:
Info + register: bit.ly/21math2025
Video part 1:
April 23, 2025 at 8:49 PM
Are you a mathematics teacher? Would you like a taste of 21st century mathematics? See the link below to join a program for teachers run by Justin Lanier @jlanier.bsky.social —stay for as long as you like, explore with friends, & get your own insights into accessible, deep questions in math
Reposted by Deep Patel
Today, I'm revisiting (some slides which I made for) a short course which I gave in December on "Geometric Functional Inequalities for Markov Chains", and I'm thinking about other ways of motivating the study of this topic. I want to have a go at re-motivating it, from scratch:
April 13, 2025 at 6:31 PM
Today, I'm revisiting (some slides which I made for) a short course which I gave in December on "Geometric Functional Inequalities for Markov Chains", and I'm thinking about other ways of motivating the study of this topic. I want to have a go at re-motivating it, from scratch:
Reposted by Deep Patel
Unsolicited writing advice, no. 96: Nothing you write is ever wasted. It’s all part of your training. That’s why you should never feel bad about the work you cut, or the hours you spend on work that stays unpublished. Every part of the process makes you a stronger writer.
April 10, 2025 at 10:27 AM
Unsolicited writing advice, no. 96: Nothing you write is ever wasted. It’s all part of your training. That’s why you should never feel bad about the work you cut, or the hours you spend on work that stays unpublished. Every part of the process makes you a stronger writer.
Reposted by Deep Patel
There's a nice symmetry to "listen, then read, then write, then tell."
April 6, 2025 at 10:47 PM
There's a nice symmetry to "listen, then read, then write, then tell."
Reposted by Deep Patel
Part 2 of the collaboration with Terence Tao on the cosmic distance ladder is now out. It covers how we first learned the distances to planets, stars, and galaxies far, far away.
youtu.be/hFMaT9oRbs4
youtu.be/hFMaT9oRbs4
The cosmic distance ladder with Terence Tao, part 2
YouTube video by 3Blue1Brown
youtu.be
February 23, 2025 at 3:00 PM
Part 2 of the collaboration with Terence Tao on the cosmic distance ladder is now out. It covers how we first learned the distances to planets, stars, and galaxies far, far away.
youtu.be/hFMaT9oRbs4
youtu.be/hFMaT9oRbs4
Reposted by Deep Patel
This semester I‘m giving „Introduction to Topological Data Analysis“ for the 3rd time. The number of registrations has grown, this year we are at 80 registrations, it‘s great to see more and more students interested in TDA!
We have lecture notes on our course webpage
ti.inf.ethz.ch/ew/courses/T...
We have lecture notes on our course webpage
ti.inf.ethz.ch/ew/courses/T...
TDA 2025 (Theory of Combinatorial Algorithms, ETH Zürich)
ti.inf.ethz.ch
February 18, 2025 at 1:24 PM
This semester I‘m giving „Introduction to Topological Data Analysis“ for the 3rd time. The number of registrations has grown, this year we are at 80 registrations, it‘s great to see more and more students interested in TDA!
We have lecture notes on our course webpage
ti.inf.ethz.ch/ew/courses/T...
We have lecture notes on our course webpage
ti.inf.ethz.ch/ew/courses/T...
Reposted by Deep Patel
Modern supply chains don't look like trade theory 101!
They involve constant border crossings, each now hit by tariffs.
Tariffs raise prices, but the more important thing they do is disrupt supply relationships.
1/
They involve constant border crossings, each now hit by tariffs.
Tariffs raise prices, but the more important thing they do is disrupt supply relationships.
1/
February 2, 2025 at 12:56 PM
Modern supply chains don't look like trade theory 101!
They involve constant border crossings, each now hit by tariffs.
Tariffs raise prices, but the more important thing they do is disrupt supply relationships.
1/
They involve constant border crossings, each now hit by tariffs.
Tariffs raise prices, but the more important thing they do is disrupt supply relationships.
1/
Reposted by Deep Patel
Quite a nice watch:
youtu.be/TE4R8bumI-Q?...
youtu.be/TE4R8bumI-Q?...
On Theoretical Foundations of Diffusion Models
YouTube video by C3 Digital Transformation Institute
youtu.be
January 6, 2025 at 9:01 PM
Quite a nice watch:
youtu.be/TE4R8bumI-Q?...
youtu.be/TE4R8bumI-Q?...
Reposted by Deep Patel
I've realised a couple of times over the break that when you're able to set aside a block of time for free research (even just e.g. a full afternoon), the benefits of having that unconstrained time are super nice in some quite specific ways.
January 3, 2025 at 9:44 PM
I've realised a couple of times over the break that when you're able to set aside a block of time for free research (even just e.g. a full afternoon), the benefits of having that unconstrained time are super nice in some quite specific ways.
Reposted by Deep Patel
Finally compiled and refined this a bit; do take a look if you're curious! A more TeX'd version is attached in the screenshots.
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
December 30, 2024 at 3:42 PM
Finally compiled and refined this a bit; do take a look if you're curious! A more TeX'd version is attached in the screenshots.
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
hackmd.io/@sp-monte-ca...
'Some Intuition Boosters'
Reposted by Deep Patel
Calling #MathSky: I'm teaching a sophomore-level class this spring called Geometry and the Arts, and I'm looking to round out my list of topics / activity ideas / materials. The goal is "Here are lots of ways that geometrical ideas come up when people are pursuing their artistic practices."
December 20, 2024 at 3:01 PM
Calling #MathSky: I'm teaching a sophomore-level class this spring called Geometry and the Arts, and I'm looking to round out my list of topics / activity ideas / materials. The goal is "Here are lots of ways that geometrical ideas come up when people are pursuing their artistic practices."
Reposted by Deep Patel
Lots of great ideas and resources in this thread!
Calling #MathSky: I'm teaching a sophomore-level class this spring called Geometry and the Arts, and I'm looking to round out my list of topics / activity ideas / materials. The goal is "Here are lots of ways that geometrical ideas come up when people are pursuing their artistic practices."
December 22, 2024 at 12:11 AM
Lots of great ideas and resources in this thread!
Reposted by Deep Patel
My book is (at last) out, just in time for Christmas!
A blog post to celebrate and present it: francisbach.com/my-book-is-o...
A blog post to celebrate and present it: francisbach.com/my-book-is-o...
December 21, 2024 at 3:23 PM
My book is (at last) out, just in time for Christmas!
A blog post to celebrate and present it: francisbach.com/my-book-is-o...
A blog post to celebrate and present it: francisbach.com/my-book-is-o...
Reposted by Deep Patel
i recently gave a talk about questions like this at a workshop organized by the amazing "learning theory alliance" (let-all.com), and several students reached out to tell me how much my talk resonated with them.
link to the slides here:
docs.google.com/presentation...
link to the slides here:
docs.google.com/presentation...
let-all-gergo
Research styles: “fast” or “slow”? Some thoughts by Gergely Neu
docs.google.com
December 17, 2024 at 12:19 PM
i recently gave a talk about questions like this at a workshop organized by the amazing "learning theory alliance" (let-all.com), and several students reached out to tell me how much my talk resonated with them.
link to the slides here:
docs.google.com/presentation...
link to the slides here:
docs.google.com/presentation...
Reposted by Deep Patel
Reposting with alt text!
November 30, 2024 at 2:39 PM
Reposting with alt text!
Reposted by Deep Patel
The Gaussian is a nice bumpy shape, but sometimes we hope for a smooth (i.e. C∞) function like the Gaussian that is 𝒂𝒍𝒔𝒐 compactly supported.
One such class of functions is called "Bump functions"
1/6
One such class of functions is called "Bump functions"
1/6
November 29, 2024 at 9:39 PM
The Gaussian is a nice bumpy shape, but sometimes we hope for a smooth (i.e. C∞) function like the Gaussian that is 𝒂𝒍𝒔𝒐 compactly supported.
One such class of functions is called "Bump functions"
1/6
One such class of functions is called "Bump functions"
1/6
Reposted by Deep Patel
Found slides by Ankur Moitra (presented at a TCS For All event) on "How to do theoretical research." Full of great advice!
My favourite: "Find the easiest problem you can't solve. The more embarrassing, the better!"
Slides: drive.google.com/file/d/15VaT...
TCS For all: sigact.org/tcsforall/
My favourite: "Find the easiest problem you can't solve. The more embarrassing, the better!"
Slides: drive.google.com/file/d/15VaT...
TCS For all: sigact.org/tcsforall/
December 13, 2024 at 8:31 PM
Found slides by Ankur Moitra (presented at a TCS For All event) on "How to do theoretical research." Full of great advice!
My favourite: "Find the easiest problem you can't solve. The more embarrassing, the better!"
Slides: drive.google.com/file/d/15VaT...
TCS For all: sigact.org/tcsforall/
My favourite: "Find the easiest problem you can't solve. The more embarrassing, the better!"
Slides: drive.google.com/file/d/15VaT...
TCS For all: sigact.org/tcsforall/
Reposted by Deep Patel
Releasing lectures and notes for an intro course on Statistical Detection & Estimation I used to teach. The core material hasn't changed (was an EE class) but it is as relevant today to AI researchers as ever before
Lecture notes
drive.google.com/file/d/1Zl_p...
YouTube
youtube.com/playlist?lis...
Lecture notes
drive.google.com/file/d/1Zl_p...
YouTube
youtube.com/playlist?lis...
December 12, 2024 at 3:44 AM
Releasing lectures and notes for an intro course on Statistical Detection & Estimation I used to teach. The core material hasn't changed (was an EE class) but it is as relevant today to AI researchers as ever before
Lecture notes
drive.google.com/file/d/1Zl_p...
YouTube
youtube.com/playlist?lis...
Lecture notes
drive.google.com/file/d/1Zl_p...
YouTube
youtube.com/playlist?lis...
Reposted by Deep Patel
As my first post on this platform, allow me to advertise the RL theory lecture notes I have been developing with Sasha Rakhlin: arxiv.org/abs/2312.16730
(shameless repost of my pinned tweet)
(shameless repost of my pinned tweet)
November 21, 2024 at 2:48 PM
As my first post on this platform, allow me to advertise the RL theory lecture notes I have been developing with Sasha Rakhlin: arxiv.org/abs/2312.16730
(shameless repost of my pinned tweet)
(shameless repost of my pinned tweet)