Uzu Lim

Uzu Lim

@uzulim.bsky.social

🍩 Postdoc at Queen Mary, working on Geometric Data Analysis πŸŽ“ Formerly at Oxford and Postech πŸ³οΈβ€βš§οΈ Trans non-binary, Any pronouns :) ✨ Also: number theory, physics, art, music https://sites.google.com/view/uzulim/main

69 Followers 74 Following 30 Posts Joined Jun 2024
9 months ago
Preview
GitHub - gdisarra/topology-degree: Demo and tutorial for the computation of the topology degree Demo and tutorial for the computation of the topology degree - gdisarra/topology-degree

πŸš€ Just released a new tool for the neuroscience and Topological Data Analysis communities!
πŸ”— github.com/gdisarra/top...

The topology-degree repository provides code and methods to quantify the topology in high-dimensional data using persistent homology.
πŸ”— doi.org/10.1371/jour...

5 2 0 0
8 months ago

@jeffgian.bsky.social Do you ever need to use spinors when doing Lattice QCD? To consider quarks, and not just gluons?

0 0 0 0
8 months ago
Preview
Topological Data Analysis of Abelian Magnetic Monopoles in Gauge Theories Motivated by recent literature on the possible existence of a second higher-temperature phase transition in Quantum Chromodynamics, we revisit the proposal that colour confinement is related to the dy...

Xavier Crean, @jeffgian.bsky.social and Biagio Lucini worked on analysing the topological data arising from lattice gauge theory, like Lattice QCD. See : arxiv.org/abs/2501.19320

I think this is one of the super meaningful applications of TDA, because the data is intrinsically topological (SU(3)).

1 0 1 0
8 months ago

If you apply the path integral approach to a discretised computer simulation of QCD, you get Lattice QCD. When doing Lattice QCD, I think you assign SU(3) elements to each edge of a 4D cubical grid. That now gives you ... some topological data. Tada..!

0 0 1 0
8 months ago

The QCD Lagrangian has a Dirac operator which makes quarks and gluons interact. With this Lagrangian, you can either write down scattering amplitudes or Feynman path integrals. My understanding is that the path integral approach still keeps the differential geometry intact.

1 0 1 0
8 months ago

I forgot what a spinor is.. a Spin group is a subset of a Clifford algebra. You need this stuff to define fermions, which are electrons and quarks (and other stuff like neutrino). To define quantum chromodynamics (QCD), you get the spin bundle and also a SU(3)-bundle and define a Lagrangian.

0 0 1 0
8 months ago

A connection 1-form is also called a gauge field. This formalises, for example, gluon fields. For a full formal account of the standard model, you still need a lot more abstract machinery. Next up is spinors.

0 0 1 0
8 months ago

A connection 1-form omega is a g-valued 1-form on P, satisfying some well-behavedness wrt G. Ehremsann connection and connection 1-form are equivalent notions, and the horizontal subspace specified in Ehresmann connection is exactly the kernel of the projection map specified by a connection 1-form.

0 0 1 0
8 months ago

An Ehresmann connection H is a sub-bundle of the tangent bundle of a principal G-bundle P->M (double bundle??) The rank of H is equal to the dimension of the base manifold M. H is chosen so that it's "horizontal" and well-behaved wrt G-action.

0 0 1 0
8 months ago

Learning gauge theory nowadays, another attempt to learn more mathematical physics

0 0 1 0
10 months ago
Post image

happy mother's day

30,500 4,100 130 118
10 months ago
Video thumbnail

"Excuse me, I'll tell you when you've finished"

4,659 614 93 58
1 year ago
Post image

A computation that took us over 8 hours now only takes 15 seconds with tropical geometry! Very excited that Oskar Henriksson will join the algebra in data analysis (AIDA) group @mpicbg.bsky.social @tudresden.bsky.social

19 3 1 1
10 months ago
Preview
Cover Learning for Large-Scale Topology Representation Classical unsupervised learning methods like clustering and linear dimensionality reduction parametrize large-scale geometry when it is discrete or linear, while more modern methods from manifold lear...

My paper with Luis Scoccola and @haharrington.bsky.social were accepted to present in ICML 2025 πŸ₯³πŸ₯³
We use gradient descent to solve a fundamental problem in computational topology: Compute a topologically good cover of a dataset. It improves Mapper and persistent homology!

arxiv.org/abs/2503.09767

17 3 0 0
10 months ago

Hi, this is my new professional account :)
I'm a mathematician working on geometry and topology of data. Recently I'm focused on Bobrowski-Skraba's universal statistical law of persistent homology, and also neuroscience applications - toroidal grid cell topology.

2 1 0 0
1 year ago

helo

0 0 1 0
1 year ago

would u accept "shittification"
also theres a math word called "sheafification"

1 0 1 0
1 year ago

me soon

1 0 0 0
1 year ago

λ”λΆˆμ–΄ μ € 본인도 upper middle class으둜 세상을 λ‹€λΆ„νžˆ λΉ„λš€μ–΄μ§€κ²Œ λ΄€λŠ”λ°, μ—¬λŸ¬ 이유둜 μ΄μ œλŠ” 쒌파둜 μ •μ²΄ν™”ν•˜κ²Œ λμ–΄μš”. 더 μ •μ§ν•˜κ³  λ°”λ₯΄κ²Œ μ‚΄ 수 있게 λμ–΄μš”.

0 0 0 0
1 year ago

μΉ΄νˆ¬μ‚¬λ‘œ μΌν• λ•Œ λ―Έκ΅­ μœ ν•™κ°„ 애듀이 λ‹€ 그런 λŠλ‚Œμ΄μ—ˆμ–΄μ„œ μ—¬λŸ¬λͺ¨λ‘œ 인상이 μ•ˆμ’‹κ²Œ λ°•νžŒ 기얡이 λ‚˜λ„€μš”.

0 0 1 0
1 year ago

he seemed viral in south korea, what's big about him?

1 0 1 0
1 year ago
Official museum photo of four small clay figurines in the form of Bactrian Camels (two humps), with details such as facial features, fur, harnesses, etc. suggested by incised lines; aranged in a line left to right from largest to smallest in side profile against orange background.
#1 Height: 7.60 cm / Width: 8 cm
#2 Height: 6 cm / Width: 5.20 cm
#3 Height: 3.81 cm / Width: ?
#4 Height: 3.30 cm / Width: 4 cm
Image Β© The Trustees of the British Museum (CC BY-NC-SA 4.0)
https://www.britishmuseum.org/collection/object/A_MAS-24
https://www.britishmuseum.org/collection/object/A_MAS-8
https://www.britishmuseum.org/collection/object/A_1902-1220-375
https://www.britishmuseum.org/collection/object/A_MAS-25

For #WorldCamelDay πŸͺ🐫:
4 figurines of Bactrian #Camels
Yotkan, Xinjiang, China, 1st–6th c. CE
Clay, H 7.6, 6.0, 3.8, 3.3 cm
British Museum collection MAS.24, MAS.8, 1902,1220.375, MAS.25:
www.britishmuseum.org/collection/o...

16 5 0 1
1 year ago
DA Rosa Lark Ascending. A pale white rose that opens with a glowing butter yellow center.

🌱Lark Ascending.
A stunning performance this season. Only more good things to come.

329 34 7 0
1 year ago

reminds me of the "The Boring Side" that the ingredient lists get written as for the "hip" companies like Oatley and Aussie

1 0 0 0
1 year ago
Preview
λ‚˜λ―ΈλΉ„μ•„ 고등법원, μ‹λ―Όμ‹œλŒ€ μ œμ •λœ 동성애 κΈˆμ§€λ²• "μœ„ν—Œ" λ‚˜λ―ΈλΉ„μ•„ λŒ€λ²•μ›μ΄ 동성애λ₯Ό λ²”μ£„ν™”ν•œ 식민지 μ‹œλŒ€μ˜ 법λ₯  두 개λ₯Ό μœ„ν—Œμ΄λΌκ³  κ²°μ •ν–ˆλ‹€.21일(ν˜„μ§€μ‹œκ°) λ‘œμ΄ν† ν†΅μ‹ μ— λ”°λ₯΄λ©΄ λ‚˜λ―ΈλΉ„μ•„ λŒ€λ²•μ›μ€ 영ꡭ μΈκΆŒλ‹¨μ²΄μ˜ 지원을 받은 μ‹œλ―Όμ‚¬νšŒλ‹¨μ²΄ HDT(Human Dignity Trust) ν™œλ™κ°€ 프리데 λ‹€μš°μ‚¬λΈŒκ°€ μ œκΈ°ν•œ ν—Œλ²•μ†Œμ›μ—μ„œ μœ„ν—Œκ²°μ •μ„ λ‚΄λ Έ

λ‚˜λ―ΈλΉ„μ•„ λŒ€λ²•μ›μ΄ 동성애λ₯Ό λ²”μ£„ν™”ν•œ 식민지 μ‹œλŒ€μ˜ 법λ₯  두 개λ₯Ό μœ„ν—Œμ΄λΌκ³  κ²°μ •ν–ˆλ‹€.
www.womennews.co.kr/news/article...

12 22 0 0
1 year ago

smel flowa every day........ lov is everywhere

1 0 0 0
1 year ago

@nyuvyu.bsky.social bee kind

1 0 1 0
1 year ago

hi angel wow...

1 0 0 0
1 year ago
Preview
β€œμΈκ°„μ΄ 가끔 엉망일지라도···차별에 λ§žμ„  싸움은 κ³„μ†λœλ‹€β€ [ν”Œλž«] μ–΄λ–€ 차별은 곡기처럼 μ‘΄μž¬ν•œλ‹€. μ–΄λŠμƒˆ μ†Œμˆ˜μžλ₯Ό μ—μ›Œμ‹Όλ‹€. μ†Œμˆ˜μžλŠ” μ°¨λ³„μ˜ μ‹œμ„ μ„ 슀슀둜 λ‚΄λ©΄ν™”ν•˜κΈ°μ— 이λ₯Έλ‹€. λͺ¨μš•에 κ°€κΉŒμš΄ 차별을 당해도 피해망상은 μ•„λ‹κΉŒ μžκΈ°κ²€μ—΄ν•˜λŠ” 식이...

μ–΄λ–€ 차별은 곡기처럼 μ‘΄μž¬ν•œλ‹€. μ–΄λŠμƒˆ μ†Œμˆ˜μžλ₯Ό μ—μ›Œμ‹Όλ‹€. μ†Œμˆ˜μžλŠ” μ°¨λ³„μ˜ μ‹œμ„ μ„ 슀슀둜 λ‚΄λ©΄ν™”ν•˜κΈ°μ— 이λ₯Έλ‹€. λͺ¨μš•에 κ°€κΉŒμš΄ 차별을 당해도 피해망상은 μ•„λ‹κΉŒ μžκΈ°κ²€μ—΄ν•˜λŠ” 식이닀. μ„£λΆˆλ¦¬ λΆ€λ‹Ήν•˜λ‹€κ³  λ§ν–ˆλ‹€κ°€λŠ” 혐였의 ν‘œμ μ΄ 될 μˆ˜λ„ μžˆλ‹€. κ²°μ½” μ‚¬μ†Œν•  수 μ—†λŠ” μ†Œμˆ˜μžμ˜ 감정을 μΊμ‹œ λ°• 홍 UCλ²„ν΄λ¦¬λŒ€ κ΅μˆ˜λŠ” β€˜λ§ˆμ΄λ„ˆ ν•„λ§μŠ€(minor feelingsΒ·μ†Œμˆ˜μ  감정)’라고 이름 λΆ™μ˜€λ‹€.
m.khan.co.kr/culture/cult...

21 18 0 0