Lecture · Chalkboard Notes

Unsolved

Open problems in mathematics

Some questions sit on the board for centuries. They look simple. They aren't.

I.

Why open problems matter

Mathematics doesn't grow outward from facts. It grows around the things it cannot prove.

  • An open problem is a tiny crack — and entire fields rush in to fill it.
  • Fermat's Last Theorem (open 358 years) gave us modern algebraic number theory.
  • The four-color theorem dragged proofs into the age of computers.
  • Failed attempts often become more useful than a hypothetical solution.

A good problem is a generator. A solved problem is a souvenir.

II.

The Millennium Prize Problems

Posted by the Clay Mathematics Institute in May 2000. Seven problems. $1,000,000 each.

P vs NP

Computation's deepest divide.

Riemann

Zeros of the zeta function.

Yang–Mills

Mass gap in gauge theory.

Navier–Stokes

Smooth flow, forever?

Birch & S-D

Elliptic curves & L-functions.

Hodge

Algebra meets topology.

The Poincaré Conjecture was solved by Grigori Perelman in 2003. He declined both the prize and the Fields Medal. Six remain.

III.

P vs NP

Is P = NP ?

If a solution can be checked in polynomial time, can it also be found in polynomial time?

  • Sudoku puzzles, protein folding, route optimization — easy to verify, brutal to solve.
  • Most mathematicians believe P ≠ NP. No one can prove it.
  • A proof either way would reshape cryptography, biology, AI.
Cook–Levin (1971) made it formal. Half a century later: nothing.
P NP NP-complete? if P = NP, the outer ring collapses inward

IV.

The Riemann Hypothesis

ζ(s) = Σ 1 / ns

All non-trivial zeros of ζ(s) lie on the critical line Re(s) = 1/2.

  • Proposed by Bernhard Riemann in 1859.
  • Equivalent to deep statements about how primes are distributed.
  • Verified for the first 1013+ zeros — none stray off the line.

"If I were to awaken after a thousand years, my first question would be: has the Riemann Hypothesis been proven?" — David Hilbert

Re(s) Im(s) Re(s) = 1/2 trivial zeros: −2, −4, −6, …

V.

Yang–Mills & the Mass Gap

A question stitched between mathematics and physics.

  • Quantum Yang–Mills theory underlies the Standard Model of particle physics.
  • Experiments imply force-carrying particles like gluons should have a positive minimum energy — a "mass gap."
  • The challenge: build a mathematically rigorous theory and prove the gap exists.
  • No one even knows how to formally define the relevant quantum field on R4.
In short: physicists use it daily; mathematicians can't agree it exists.

VI.

Navier–Stokes Existence & Smoothness

The equations that describe every fluid you've ever seen — water, air, blood, weather.

∂u/∂t + (u·∇)u = −∇p + ν Δu
  • Given smooth initial conditions in 3D, do solutions stay smooth — or can a vortex blow up to infinity in finite time?
  • Engineers compute with these equations daily, trusting them implicitly.
  • If singularities can form, our model of turbulence is incomplete at its very foundation.

A river of equations. Nobody knows whether it floods.

VII.

Birch & Swinnerton-Dyer

y2 = x3 + ax + b

The number of rational points on an elliptic curve is governed by the behavior of an associated L-function at s = 1.

  • Conjectured in the 1960s using early computer experiments at Cambridge.
  • Connects three worlds: arithmetic, complex analysis, geometry.
  • Elliptic curves powered Wiles' proof of Fermat's Last Theorem and underlie modern cryptography.
y² = x³ − x + 1 P Q P + Q

VIII.

The Hodge Conjecture

The most abstract of the seven. The hardest to even state casually.

  • On a smooth complex projective variety, certain topological features (Hodge classes) should always come from algebraic geometry.
  • Translation: shapes you can see with topology should always be cut out by polynomial equations.
  • It is a bridge — promising that two languages describe the same continent.
  • Known in low dimensions and for specific cases. Open in general.

If true: a deep unity. If false: a strange and useful asymmetry.

IX.

The Twin Prime Conjecture

Are there infinitely many primes p such that p + 2 is also prime?

(3, 5), (5, 7), (11, 13), (17, 19), … (1016 + 1, …) ?

  • Easy to state. Proposed in some form by de Polignac in 1849.
  • 2013 — Yitang Zhang, an unknown lecturer, proved infinitely many prime pairs differ by at most 70,000,000.
  • Within months, the Polymath project shrank the bound to 246.
  • From 70 million to 246 in a year. From 246 to 2: still nobody.

X.

The Collatz Conjecture

if n is even: n → n/2    ·    if n is odd: n → 3n + 1

Pick any positive integer. Apply the rule. Repeat.

  • Conjecture: every starting n eventually reaches 1.
  • Verified for every n < 2.95 × 1020.
  • Paul Erdős: "Mathematics may not be ready for such problems."
  • Looks like a homework exercise. Has resisted proof for 90 years.
1 2 4 8 — 16 5 10 — 32 3 20 21 does every branch fall to 1?

XI.

Goldbach's Conjecture

Every even integer greater than 2 is the sum of two primes.

4 = 2 + 2   ·   6 = 3 + 3   ·   100 = 3 + 97   ·   1018 = …
  • Proposed in a 1742 letter from Christian Goldbach to Leonhard Euler.
  • Verified by computer up to 4 × 1018.
  • Helfgott (2013) proved the weak Goldbach: every odd integer > 5 is a sum of three primes.
  • The strong version — sum of two primes — has stood untouched for nearly three centuries.

XII.

Where to read & watch

A few directions if any of these problems caught you.

  • The Music of the Primes — Marcus du Sautoy
  • Prime Obsession — John Derbyshire (the Riemann Hypothesis)
  • The Millennium Problems — Keith Devlin
  • Clay Mathematics Institute — official problem statements (claymath.org)
  • Quanta Magazine — superb reporting on number theory breakthroughs
YouTube: Riemann Hypothesis explained → YouTube: P vs NP problem →

"In mathematics the art of asking questions is more valuable than solving them." — Cantor

1 / 13
← → space · click anywhere