
AI Roadmap


Fermat’s Little Theorem used in RSA Encryption.
This guy demo how modern math can link / prove from many angles : Euler Totient φ number, Group theory with Lagrange Theorem , using the Set function Surjective, Injective, Bijective fundamental concept technique …
Also can prove by ancient Chinese in 200 BCE from Chinese Remainder Theorem (CRT)…
条条大路通罗马。
Beauty of modern maths is the REAL Maths study, not the UK/China computational maths.
“Fermat” is my Alma mater lycée (Classe Prepa, University years 1&2 ) in France (Toulouse, AirBus HQ City). He was a 17CE judge cum amateur mathematician invented Number Theory, Probability (co- with Pascal), Analytic Geometry (co- with Descartes), and Physics (Refraction Law.)
Great video, quite different from others on Riemann Zeta Function for non-trivial zero.
Must view at least 2 times to understand the final part with Cauchy Residual Theorem.
Python GUI

You can ask Gemini to explain the X-ENS oral test, remember to remind it to “说人话”, treat you the student as an “idiot” , so it wld explain from ABC basic using layman analogy 🙂
Explain this math oral test of X-ENS in Linear Algebra, please 说人话 the concepts here : trace, exp ?
That’s the new pedagogical method of AI-approach e-Learning. A Swedish A-level student learns AI from zero ground to being hired as the Open-AI PhD-level researcher in 5 years.
Solution by Grok: his method with (& without) Jordan Form.
To get Fields Medal is ½ by luck (according to 陈省身 ) – ie to continue the final leg work of others , then 临门一脚 , 踢进球🏀!
除非是天才,得Fields 奖是靠
一半天份,一半运气.!
eg:
王虹是运气,遇到 合作 Zahl (thru Terence Tao) 和 法国/MIT.
天才 “韦神”没有王虹的 “幸运” 碰到 贵人。
武侠的第一高手 郭靖, 杨过,张无忌, 令狐冲,张君宝(三丰) 。。。都是 在江湖闯荡,碰到贵人师傅(少林 扫地僧, 丐帮主洪七公,东邪 黄药师, 周不通,) /女友or 爱人 (小龙女,黄蓉,Iran明教教主 小昭/ 护法 金狮王,。。。 )
Those with O level Set Theory aka New Maths will cope better in Uni pure math aka Abstract Math.
This is Beautiful if you know IDEAL : the creation by Dedekind for unique factorisation of integers to defy Gauss Fundamental Theorem of Algebra.
Next chapter using Elliptic Curve (which is used to proof Fermat Last Theorem) is more advanced even the video’s author also not fully understand all “why”.
Claude Code, hosting, front-end, backend, password,resend, GitHub …
In Maths education:
Level 1 (Primary School) : Numbers
Level 2 (Sec School: O- & A-level) : Functions
Level 3 (Undergrad): Structure Abstraction (Group, Ring, Field, …)
Level 4 (Grad School = MSc, PhD) : Category = Abstraction of Abstractions
王虹 on Kakeya Conjecture : proved from
dim n= 2 (already known)
to
n=2.1, 2.2, 2.3, …2.9, 3
by induction.
Tonnetz (p,r,L) transformations from C Maj to Cm, to Em, to Am, etc .
This is “Group” Transformation in disguise in 1800 AD, 20 Years before Galois invented Group Theory ?
Gemini 3 confirms my suspicion above: Tonnetz by Euler is a Group Action in disguise.
——-Gemini 3 Reply —-
It is a brilliant observation to connect Euler’s Speculum Musicum (the basis for the Tonnetz) with the formal birth of Group Theory. While Euler wasn’t intentionally writing down group axioms in the late 1700s, he was absolutely working with the algebraic structure that Galois would later formalize.
Technically, you are right: Euler’s work on musical intervals and transformations describes a group in everything but name.
The Mathematical “Disguise”
Leonhard Euler’s Tentamen novae theoriae musicae (1739) and his later work on the Tonnetz (the “tone web”) utilized what we now call a Group Action.
Here is how Euler’s work pre-dated Galois by using group-like properties:
* *Closure* : Euler recognized that moving by a perfect fifth or a major third always landed you on another note within the pitch-class space.
* *Inverses* : He understood that moving “up” a fifth could be undone by moving “down” a fifth.
* *Commutativity* : On a 2D Tonnetz, moving a Major Third then a Perfect Fifth gets you to the same place as a Fifth then a Major Third. This describes an *Abelian Group* .
*Why it wasn’t “Group Theory” yet* ?
While Euler was using the structure, he lacked the Group abstraction.
*The Neo-Riemannian Connection*
The specific transformations you mentioned—P (Parallel), R (Relative), and L (Leittonwechsel)—were actually formalized much later by Hugo Riemann and modern theorists like David Lewin.
However, these transformations form the *PRL Group* . If we look at them through Euler’s lens:
* They are *involutions* 内卷: Applying P twice (P²) returns you to the original chord, meaning P = P-¹.
* They operate on a set of 24 major and minor triads.
Euler was essentially navigating the Cayley Graph of a group before Arthur Cayley or Évariste Galois were even born. He had the “map” (the Tonnetz), but Galois provided the “laws of motion” that explained why the map worked.
1584 AD 朱载堉 (明朝 太子) invented 12 tones (copied by French Jesuit 白晋 in China 康熙 Court -> informed Germany Lebniz-> taught Sebastian Bach who implemented it in first modern piano today).
Any closed system of musical transformations has implicit group structure (just like the circle of fifths is ℤ/12ℤ).
Five stages (levels) of AI:
Not always true to interchange integral ∫ with summation Σ .
Markov Chain (Probability)
its unique feature :
it settles ultimately to a steady state.
This was one interesting “Colle” oral test in Analysis @ Maths Sup.

Prove by Calculus: MVT


Beautiful American math: using 1) matrix in Linear Algebra, 2) isomorphism, 3) commutative diagram.
Wang Hong, 2026 Fields Medal ?

She was educated in:
中: 北大数学 Bachelor
F: Ecole Polytechnique (Diplôme, Architecture Engr)+ Paris U (MSc Maths)
A: MIT (PhD. Maths)
Lens in CATEGORY Theory implemented in Haskell & Python codes


no coding, just drag & drop