Simon problems

Not to be confused with Simon's problem in computational complexity.

In mathematics, the Simon problems (or Simon's problems) are a series of fifteen questions posed in the year 2000 by Barry Simon, an American mathematical physicist. Inspired by other collections of mathematical problems and open conjectures, such as the famous list by David Hilbert, the Simon problems concern quantum operators. Eight of the problems pertain to anomalous spectral behavior of Schrödinger operators, and five concern operators that incorporate the Coulomb potential.

In 2014, Artur Avila won a Fields Medal for work including the solution of three Simon problems. Among these was the problem of proving that the set of energy levels of one particular abstract quantum system was in fact the Cantor set, a challenge known as the "Ten Martini Problem" after the reward that Mark Kac offered for solving it.

The 2000 list was a refinement of a similar set of problems that Simon had posed in 1984.

Context

Background definitions for the "Coulomb energies" problems:

• ${\mathcal {H}}_{f}^{(N)}$ is the space of functions on $L^{2}({\mathcal {R}}^{2N};{\mathcal {C}}^{3N})$ which are antisymmetric in spin and space.
• The Hamiltonian is ${\mathcal {H}}(N,Z):=\sum _{i=1}^{N}(-\Delta _{i}-{\frac {Z}{x_{i}}})+\sum _{i<j}{\frac {1}{|x_{i}-x_{j}|}}$.
• We have $E(N,Z):=\min _{{\mathcal {H}}_{f}}H(N,Z)$.
• We define $N_{0}(Z)$ to be the smallest value of $E(N+j,Z)=E(N,Z)$ for all positive integers $j$; it is known that such a number always exists and is always between $Z$ and $2Z$, inclusive.

The 1984 list

Simon listed the following problems in 1984:

No.	Short name	Statement	Status	Year solved
1st	(a) Almost always global existence for Newtonian gravitating particles (b) Existence of non-collisional singularities in the Newtonian N-body problem	(a) Prove that the set of initial conditions for which Newton's equations fail to have global solutions has measure zero. (b) Show that there are non-collisional singularities in the Newtonian N-body problem for some N and suitable masses.	Open as of 1984.	?
2nd	(a) Ergodicity of gases with soft cores (b) Approach to equilibrium (c) Asymptotic abelianness for the quantum Heisenberg dynamics	(a) Find repulsive smooth potentials for which the dynamics of N particles in a box (with, e.g., smooth wall potentials) is ergodic. (b) Use the above scenario to justify that large systems with forces that are attractive at suitable distances approach equilibrium, or find an alternate scenario that does not rely on strict ergodicity in finite volume.	Open as of 1984. Sinai once proved that the hard sphere gas is ergodic, but no complete proof has appeared except for the case of two particles, and a sketch for three, four, and five particles.	?
3rd	Turbulence and all that	Develop a comprehensive theory of long-time behavior of dynamical systems, including a theory of the onset of and of fully developed turbulence.	?	?
4th	(a) Fourier's heat law (b) Kubo's formula	(a) Find a mechanical model in which a system of size $L$ with temperature difference $\Delta T$ between its ends has a rate of heat temperature that goes as $L^{-1}$ in the limit $L\to \infty$. (b) Justify Kubo's formula in a quantum model or find an alternate theory of conductivity.	?	?
5th	(a) Exponential decay of $v=2$ classical Heisenberg correlations (b) Pure phases and low temperatures for the $v\geq 3$ classical Heisenberg model (c) GKS for classical Heisenberg models (d) Phase transitions in the quantum Heisenberg model	(a) Consider the two-dimensional classical Heisenberg model. Prove that for any beta, correlations decay exponentially as distance approaches infinity. (b) Prove that, in the $D=3$ model at large beta and at dimension $v\geq 3$, the equilibrium states form a single orbit under $SO(3)$, which is the sphere. (c) Let $f$ and $g$ be finite products of the form $(\sigma _{\alpha }\cdot \sigma _{\gamma })$ in the $D=3$ model. Is it true that $<fg>_{\Lambda ,\beta }\geq <f>_{\Lambda ,\beta }<g>_{\Lambda ,\beta }$ ? (d) Prove that for $v\geq 3$ and large beta, the quantum Heisenberg model has long range order.	?	?
6th	Explanation of ferromagnetism	Verify the Heisenberg picture of the origin of ferromagnetism (or an alternative) in a suitable model of a realistic quantum system.	?	?
7th	Existence of continuum phase transitions	Show that for suitable choices of pair potential and density, the free energy is non-$C^{1}$ at some beta.	?	?
8th	(a) Formulation of the renormalization group (b) Proof of universality	(a) Develop mathematically precise renormalization transformations for $v$-dimensional Ising-type systems. (b) Show that critical exponents for Ising-type systems with nearest neighbor coupling but different bond strengths in the three directions are independent of ratios of bond strengths.	?	?
9th	(a) Asymptotic completeness for short-range N-body quantum systems (b) Asymptotic completeness for Coulomb potentials	(a) Prove that $\oplus ~{\text{Ran}}~\Omega _{a}^{+}=L^{2}(X)$. (b) Suppose $v=3,V_{ij}(x)=e_{ij}|x|^{-1}$. Prove that $\oplus ~{\text{Ran}}~\Omega _{a}^{D,+}=L^{2}(X)$.	Open as of 1984.	?
10th	(a) Monotonicity of ionization energy (b) The Scott correction (c) Asymptotic ionization (d) Asymptotics of maximal ionized charge (e) Rate of collapse of Bose matter	(a) Prove that $(\Delta E)(N-1,Z)\geq (\Delta E)(N,Z)$. (b) Prove that $\lim _{Z\to \infty }(E(Z,Z)-e_{TF}Z^{7/3})/Z^{2}$ exists and is the constant found by Scott. (c) Find the leading asymptotics of $(\Delta E)(Z,Z)$. (d) Prove that $\lim _{Z\to \infty }N(Z)/Z=1$. (e) Find suitable $C_{1},C_{2},\alpha$ such that $-C_{1}N^{\alpha }\leq {\tilde {E}}_{B}(N,N;1)\leq C_{2}N^{\alpha }$.	?	?
11th	Existence of crystals	Prove a suitable version of the existence of crystals (e.g. there is a choice of minimizing configurations that converge to some infinite lattice configuration).	?	?
12th	(a) Existence of extended states in the Anderson model (b) Diffusive bound on "transport" in random potentials (c) Smoothness of $k(E)$ through the mobility edge in the Anderson model (d) Analysis of the almost Mathieu equation (e) Point spectrum in a continuous almost periodic model	(a) Prove that in $v\geq 3$ and for small $\lambda$ that there is a region of absolutely continuous spectrum of the Anderson model, and determine whether this is false for $v=2$. (b) Prove that ${\text{Exp}}(\delta _{0},(e^{itH}{\vec {N}}e^{-itH})^{2}\delta _{0})\leq c(1+|t|)$ for the Anderson model, and more general random potentials. (c) Is $k(E)$, the integrated density of states, a $C^{\infty }$ function in the Anderson model at all couplings? (d) Verify the following for the almost Mathieu equation: If $\alpha$ is a Liouville number and $\lambda \neq 0$, then the spectrum is purely singular continuous for almost all $\theta$. If $\alpha$ is a Roth number and $|\lambda |<2$, then the spectrum is purely absolutely continuous for almost all $\theta$. If $\alpha$ is a Roth number and $|\lambda |>2$, then the spectrum is purely dense pure point. If $\alpha$ is a Roth number and $|\lambda |=2$, then $\sigma (h)$has Lebesgue measure zero and the spectrum is purely singular continuous. (e) Show that $-{\frac {d^{2}}{dx^{2}}}+\lambda \cos(2\pi x)+\mu \cos(2\pi \alpha x+\theta )$ has some point spectrum for suitable $\alpha ,\lambda ,\mu$ and almost all $\theta$.	?	?
13th	Critical exponent for self-avoiding walks	Let $D(n)$ be the mean displacement of a random self-avoiding walk of length $n$. Show that $v:=\lim _{n\to \infty }n^{-1}\ln D(n)$ is ${\frac {1}{2}}$ for dimension at least four and greater otherwise.	?	?
14th	(a) Construct QCD (b) Renormalizable QFT (c) Inconsistency of QED (d) Inconsistency of $\varphi _{4}^{4}$	(a) Give a precise mathematical construction of quantum chromodynamics. (b) Construct a nontrivial quantum field theory that is renormalizable but not superrenormalizable. (c) Prove that QED is not a consistent theory. (d) Prove that a nontrivial $\varphi _{4}^{4}$ theory does not exist.	?	?
15th	Cosmic censorship	Formulate and then prove or disprove a suitable version of cosmic censorship.	?	?

The 2000 list

The Simon problems as listed in 2000 (with original categorizations) are:

No.	Short name	Statement	Status	Year solved
Quantum transport and anomalous spectral behavior
1st	Extended states	Prove that the Anderson model has purely absolutely continuous spectrum for $v\geq 3$ and suitable values of $b-a$ in some energy range.	?	?
2nd	Localization in 2 dimensions	Prove that the spectrum of the Anderson model for $v=2$ is dense pure point.	?	?
3rd	Quantum diffusion	Prove that, for $v\geq 3$ and values of $|b-a|$ where there is absolutely continuous spectrum, that $\sum _{n\in \mathbb {Z} ^{\nu }}n^{2}|e^{itH}(n,0)|^{2}$ grows like $ct$ as $t\to \infty$.	Solved by Puig (2003).	2003
4th	Ten Martini problem	Prove that the spectrum of $h_{\alpha ,\lambda ,\theta }$ is a Cantor set (that is, nowhere dense) for all $\lambda \neq 0$ and all irrational $\alpha$.	?	?
5th		Prove that the spectrum of $h_{\alpha ,\lambda ,\theta }$ has measure zero for $\lambda =2$ and all irrational $\alpha$.	Solved by Avila and Krikorian (2003).	2003
6th		Prove that the spectrum of $h_{\alpha ,\lambda ,\theta }$ is absolutely continuous for $\lambda =2$ and all irrational $\alpha$.	?	?
7th		Do there exist potentials $V(x)$ on $[0,\infty )$ such that $|V(x)|\leq C|x|^{{\frac {1}{2}}+\varepsilon }$ for some $\varepsilon$ and such that $-{\frac {d^{2}}{dx^{2}}}+V$ has some singular continuous spectrum?	Essentially solved by Denisov (2003) with only $L^{2}$ decay. Solved entirely by Kiselev (2005).	2003, 2005
8th		Suppose that $V(x)$ is a function on $\mathbb {R} ^{\nu }$ such that $\int |x|^{-\nu +1}|V(x)|^{2}d^{\nu }x<\infty$, where $\nu \geq 2$. Prove that $-\Delta +V$ has absolutely continuous spectrum of infinite multiplicity on $[0,\infty )$.	?	?
Coulomb energies
9th		Prove that $N_{0}(Z)-Z$ is bounded for $Z\to \infty$.	?	?
10th		What are the asymptotics of$(\delta E)(Z):=E(Z,Z-1)-E(Z,Z)$ for $Z\to \infty$?	?	?
11th		Make mathematical sense of the nuclear shell model.	?	?
12th		Is there a mathematical sense in which one can justify current techniques for determining molecular configurations from first principles?	?	?
13th		Prove that, as the number of nuclei approaches infinity, the ground state of some neutral system of molecules and electrons approaches a periodic limit (i.e. that crystals exist based on quantum principles).	?	?
Other problems
14th		Prove that the integrated density of states $k(E)$ is continuous in the energy.	?	?
15th	Lieb-Thirring conjecture	Prove the Lieb-Thirring conjecture on the constants $L_{\gamma ,\nu }$ where $\nu =1,{\frac {1}{2}}<\gamma <{\frac {3}{2}}$.	?	?

See also

• Almost Mathieu operator
• Lieb–Thirring inequality

External links

• "Simon's Problems". MathWorld. Retrieved 2018-06-13.