Compilation··3:08·Dynamical systems and ergodic theory
Birkhoff Conjecture, Zero Entropy and 7 More in Dynamical Systems and Ergodic Theory
Watch on YouTubeAbout the problem →
About this video
9 results from OpenAI's public mathematics catalogue, one per chapter. Each chapter names the problem, shows the real objects, and states what OpenAI claims; none of it is peer reviewed yet.
Chapters:
0:00 Banach's Lebesgue Spectrum Problem
0:21 Rokhlin's Multiple Mixing Problem
0:42 Birkhoff Conjecture
1:04 Entropy-Rate Dimension Formula for Self-Similar Measures
1:26 Permanence Conjecture
1:46 Shub's Entropy Conjecture
2:06 Zero Entropy, No Smooth Model
2:27 Bernoulli Convolutions
2:47 Pointwise Multiple Ergodic Averages
Banach's Lebesgue Spectrum Problem: Banach's simple Lebesgue-spectrum problem, in its probability-preserving form, asks for a measure-preserving map T and one pattern f whose shuffled copies f, f∘T, f∘T², … (and backwards) are perfectly independent…
Explained: gradientdamage.com/p/144
Paper: github.com/openai/math/blob/main/preprints/A-smooth-three-tor…
Rokhlin's Multiple Mixing Problem: Stir a cup long enough and any blob of milk ends up spread evenly: that is mixing, where two sets become independent.
Explained: gradientdamage.com/p/145
Paper: github.com/openai/math/blob/main/preprints/Rokhlins-multiple-…
Birkhoff Conjecture: Birkhoff's billiards conjecture asks: if a ball bouncing inside a smooth convex table always keeps to smooth curves (caustics), must the table be an ellipse? OpenAI claims a proof of the near-boundary version: a smooth…
Explained: gradientdamage.com/p/147
Paper: github.com/openai/math/blob/main/preprints/Rigidity-of-Smooth…
Entropy-Rate Dimension Formula for Self-Similar Measures: How big is a fractal that overlaps itself? For a self-similar measure on the line, the expected answer is: dimension = min(1, entropy rate / Lyapunov exponent), where the entropy rate measures how many genuinely…
Explained: gradientdamage.com/p/148
Paper: github.com/openai/math/blob/main/preprints/The-entropy-rate-d…
Permanence Conjecture: In a chemical reaction network where every reaction can eventually be undone through a loop of other reactions (weakly reversible), with mass-action kinetics, can a chemical still die out or blow up? The permanence…
Explained: gradientdamage.com/p/149
Paper: github.com/openai/math/blob/main/preprints/Uniform-Permanence…
Shub's Entropy Conjecture: Shub's entropy conjecture (1974) says that if a map's action on homology grows, the map must be chaotic: its topological entropy should be at least the logarithm of the homological spectral radius.
Explained: gradientdamage.com/p/151
Paper: github.com/openai/math/blob/main/preprints/A-C1-Counterexampl…
Zero Entropy, No Smooth Model: Some measure-preserving maps have zero entropy: no chaos at all.
Explained: gradientdamage.com/p/152
Paper: github.com/openai/math/blob/main/preprints/A-finite-entropy-s…
Bernoulli Convolutions: Flip a fair coin forever and add ±1, ±r, ±r², ...
Explained: gradientdamage.com/p/153
Paper: github.com/openai/math/blob/main/preprints/Arithmetic-classif…
Pointwise Multiple Ergodic Averages: Stir a cube of paint with a mixing map and any blob spreads evenly through it.
Explained: gradientdamage.com/p/154
Paper: github.com/openai/math/blob/main/preprints/Pointwise-Multiple…
The animations illustrate the results; they are not proofs.
Gradient Damage: mathematics you can see, equations you can hear.
youtube.com/@gradientdamage
Chapters
One problem per chapter. Each name links to the result's own page, each time to that moment on YouTube.
- Banach's Lebesgue Spectrum Problem Dynamical systems and ergodic theory
- Rokhlin's Multiple Mixing Problem Dynamical systems and ergodic theory
- Birkhoff Conjecture Dynamical systems and ergodic theory
- Entropy-Rate Dimension Formula for Self-Similar Measures Dynamical systems and ergodic theory
- Permanence Conjecture Dynamical systems and ergodic theory
- Shub's Entropy Conjecture Dynamical systems and ergodic theory
- Zero Entropy, No Smooth Model Dynamical systems and ergodic theory
- Bernoulli Convolutions Dynamical systems and ergodic theory
- Pointwise Multiple Ergodic Averages Dynamical systems and ergodic theory
Chapter 1: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 144
Banach’s simple Lebesgue-spectrum problem
Resolves the probability-preserving form of Banach's simple Lebesgue-spectrum problem within smooth dynamics. A smooth volume-preserving diffeomorphism of the standard-volume three-torus has simple Lebesgue spectrum on its entire complex mean-zero L2 space: the bilateral iterates of one real observable form an orthonormal basis of that space. (Lean)
Chapter 2: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 145·from 0:21
Rokhlin’s multiple-mixing problem
Proves that every invertible mixing probability-preserving transformation is mixing of all finite orders, resolving Rokhlin's multiple-mixing problem for a single transformation. Correlations among any finite collection of measurable sets converge to the product of their measures whenever all pairwise time separations diverge. (Lean)
Chapter 3: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 147·from 0:42
The near-boundary Birkhoff conjecture
Resolves the near-boundary Birkhoff conjecture for smooth strictly convex planar billiards of positive curvature. Such a billiard is an ellipse whenever a full grazing annulus is continuously foliated by individually invariant essential curves. A continuous physical collar of smooth closed convex caustics also suffices.
Chapter 4: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 148·from 1:04
The entropy-rate dimension formula for self-similar measures
For every self-similar measure on the line generated by finitely many contracting similarities, proves dimHμ=min {1,hRW/χ}, where hRW is the entropy rate of random composed maps and χ the average logarithmic contraction. This resolves the entropy-rate dimension conjecture without a separation assumption, allowing exact overlaps and unequal contraction ratios. (Lean)
Chapter 5: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 149·from 1:26
Classwise permanence for weakly reversible mass-action systems
Proves the permanence conjecture for every finite weakly reversible mass-action system with fixed positive rate constants. Every positive stoichiometric compatibility class, even an unbounded one, has a common compact convex forward-invariant absorbing set. All positive trajectories in that class therefore eventually share positive lower and finite upper concentration bounds. (Lean)
Chapter 6: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 151·from 1:46
A C1 counterexample to the entropy conjecture
Constructs a noninvertible C1 self-map of a compact smooth manifold with zero topological entropy but eigenvalue 2 on second homology. This disproves the homological entropy lower bound for general C1 self-maps: homological growth need not force positive orbit complexity. (Lean)
Chapter 7: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 152·from 2:06
Zero entropy does not guarantee a smooth positive-volume model
Constructs a zero-entropy ergodic invertible transformation of a standard nonatomic probability space that is not measurably conjugate to any C∞ diffeomorphism preserving a strictly positive smooth probability density on a compact finite-dimensional manifold. One example rules out every finite dimension. (Lean)
Chapter 8: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 153·from 2:27
Arithmetic classification and non-Pisot singularity for Bernoulli convolutions
Classifies singular and absolutely continuous unbiased Bernoulli convolutions for every λ ∈ (0,1) by an infinite, one-sided approximation condition using explicit finite sets of algebraic units. It also proves singularity at reciprocals of every quartic Salem number in (1,2), giving examples beyond reciprocal Pisot parameters.
Chapter 9: about the problem
Dynamical systems and ergodic theory·OpenAI Math·Result 154·from 2:47
Pointwise multiple ergodic averages for mixing transformations
Proves almost-everywhere convergence of consecutive multiple ergodic averages of every finite length for invertible mixing probability-preserving transformations. For each fixed tuple of bounded functions, the limit is the product of their integrals, along all positive averaging lengths. No mixing rate or standardness assumption on the probability space is required.
Topics
Banach problemRokhlin problemBirkhoff conjectureself-similar measurespermanence conjectureentropy conjecturesmooth realization problemBernoulli convolutionsergodic theoryDynamical systems and ergodic theoryOpenAI mathAI proofAI mathematicsmathsGradient Damage