+ [*Partial Answer of How Transformers Learn Automata*](https://arxiv.org/abs/2504.XXXX)
+ **Tiantian Zhang**
-
CVPR 2016
-
+ We study the connection between representation theory, Fourier analysis, and how Transformers simulate finite automata, with applications to modular reasoning.
+
+
+
+---
-[Deep Residual Learning for Image Recognition](https://openaccess.thecvf.com/content_cvpr_2016/papers/He_Deep_Residual_Learning_CVPR_2016_paper.pdf)
+# đź“ť Publications
+
+
+
+
arXiv 2025
+
+
+
+
+ [*Partial Answer of How Transformers Learn Automata*](https://arxiv.org/abs/2504.XXXX)
+ **Tiantian Zhang**
-**Kaiming He**, Xiangyu Zhang, Shaoqing Ren, Jian Sun
+ We study the connection between representation theory, Fourier analysis, and how Transformers simulate finite automata, with applications to modular reasoning.
+
+
+
+
+
+
Working Draft 2025
+
+
+
+
+ [*Rank-Induced PL Mirror Descent*](#)
+ **Tiantian Zhang**, Noah [Last Name], Jieru Zuo
-# đź“– Educations
-- *2021.09 - 2025.05 Columbia University*.
+ We propose a new mirror descent algorithm for the Plackett–Luce model using a Hessian-based preconditioner, achieving improved regret bounds under the rank–distributional equivalence framework.
+
+
+# đź“– Education
+- **Columbia University** — *B.A. in Computer Science* (Sep 2021 – Dec 2025)
+ Coursework: Convex Optimization, Statistical Learning, Algorithms, Stochastic Models.
+ Honors: Dean’s List, Distinguished New Student Scholarship.
From 749bf8ea0d551a2962e04aeeed50256def9e5e0c Mon Sep 17 00:00:00 2001
From: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
Date: Wed, 17 Sep 2025 03:36:18 -0400
Subject: [PATCH 04/20] Remove news section from about.md
Signed-off-by: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
---
_pages/about.md | 6 ------
1 file changed, 6 deletions(-)
diff --git a/_pages/about.md b/_pages/about.md
index 444d70e279f..a7909cdd02b 100644
--- a/_pages/about.md
+++ b/_pages/about.md
@@ -20,12 +20,6 @@ redirect_from:
Hi! I’m **Tiantian Zhang**, an undergraduate student in Computer Science at Columbia University (Class of 2025).
My research interests span **machine learning theory**, **optimization**, and **quantitative finance** — especially Fourier-based methods for Transformers, time-series modeling, and algorithmic trading strategies.
----
-
-# 🔥 News
-- *2025.04* — Preprint on **Fourier Analysis in Proof Complexity and Transformers** posted to arXiv.
-- *2024.09* — Quantitative Research Intern at **Cantor Fitzgerald / BGC Group**, New York.
-- *2024.03* — Participated in the **MIT Premier Research Seminar** in Optimization and Systems.
---
From 1d91484faae9090b38b515685201ecab7634791b Mon Sep 17 00:00:00 2001
From: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
Date: Wed, 17 Sep 2025 03:40:40 -0400
Subject: [PATCH 05/20] Revise about page and remove publication section
Signed-off-by: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
---
_pages/about.md | 22 ++--------------------
1 file changed, 2 insertions(+), 20 deletions(-)
diff --git a/_pages/about.md b/_pages/about.md
index a7909cdd02b..8e216eb5b43 100644
--- a/_pages/about.md
+++ b/_pages/about.md
@@ -18,27 +18,9 @@ redirect_from:
Hi! I’m **Tiantian Zhang**, an undergraduate student in Computer Science at Columbia University (Class of 2025).
-My research interests span **machine learning theory**, **optimization**, and **quantitative finance** — especially Fourier-based methods for Transformers, time-series modeling, and algorithmic trading strategies.
+My research interests span **machine learning theory**, **combinatorics optimization**. During my undergrad study, I am fortunate to have worked with Yuri Faenza, Daniel Hsu.
----
-
-# đź“ť Publications
-
-
-
-
arXiv 2025
-
-
-
-
- [*Partial Answer of How Transformers Learn Automata*](https://arxiv.org/abs/2504.XXXX)
- **Tiantian Zhang**
-
- We study the connection between representation theory, Fourier analysis, and how Transformers simulate finite automata, with applications to modular reasoning.
-
-
-
---
# đź“ť Publications
@@ -66,7 +48,7 @@ My research interests span **machine learning theory**, **optimization**, and **
[*Rank-Induced PL Mirror Descent*](#)
- **Tiantian Zhang**, Noah [Last Name], Jieru Zuo
+ **Tiantian Zhang**, Jieru Zuo , Daniel Hsu
We propose a new mirror descent algorithm for the Plackett–Luce model using a Hessian-based preconditioner, achieving improved regret bounds under the rank–distributional equivalence framework.
From de33af40348942741076bfe5219a412692412780 Mon Sep 17 00:00:00 2001
From: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
Date: Wed, 17 Sep 2025 03:41:17 -0400
Subject: [PATCH 06/20] Update badge text from 'Working Draft 2025' to 'In
submission into JMLR'
Signed-off-by: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
---
_pages/about.md | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/_pages/about.md b/_pages/about.md
index 8e216eb5b43..118431faf86 100644
--- a/_pages/about.md
+++ b/_pages/about.md
@@ -42,7 +42,7 @@ My research interests span **machine learning theory**, **combinatorics optimiza
-
Working Draft 2025
+
In submission into JMLR
From 3f1f4858edd2e7aeb1697016ac28733f7d98b800 Mon Sep 17 00:00:00 2001
From: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
Date: Wed, 17 Sep 2025 03:45:16 -0400
Subject: [PATCH 07/20] Update about me section and paper links
Signed-off-by: Tiantian Zhang <85858376+crystal-cheung@users.noreply.github.com>
---
_pages/about.md | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/_pages/about.md b/_pages/about.md
index 118431faf86..582138e16ca 100644
--- a/_pages/about.md
+++ b/_pages/about.md
@@ -18,7 +18,7 @@ redirect_from:
Hi! I’m **Tiantian Zhang**, an undergraduate student in Computer Science at Columbia University (Class of 2025).
-My research interests span **machine learning theory**, **combinatorics optimization**. During my undergrad study, I am fortunate to have worked with Yuri Faenza, Daniel Hsu.
+My research interests span **machine learning theory**, **combinatorics optimization**. During my undergrad study, I am fortunate to work with Prof. Yuri Faenza, Prof. Daniel Hsu.
---
@@ -32,7 +32,7 @@ My research interests span **machine learning theory**, **combinatorics optimiza
