skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.


This content will become publicly available on February 4, 2026

Title: Randomized Runge-Kutta-Nyström methods for unadjusted Hamiltonian and kinetic Langevin Monte Carlo
We introduce 5 / 2 5/2 - and 7 / 2 7/2 -order L 2 L^2 -accurate randomized Runge-Kutta-Nyström methods, tailored for approximating Hamiltonian flows within non-reversible Markov chain Monte Carlo samplers, such as unadjusted Hamiltonian Monte Carlo and unadjusted kinetic Langevin Monte Carlo. We establish quantitative 5 / 2 5/2 -order L 2 L^2 -accuracy upper bounds under gradient and Hessian Lipschitz assumptions on the potential energy function. The numerical experiments demonstrate the superior efficiency of the proposed unadjusted samplers on a variety of well-behaved, high-dimensional target distributions.  more » « less
Award ID(s):
2111224
PAR ID:
10627750
Author(s) / Creator(s):
;
Publisher / Repository:
American Mathematical Society
Date Published:
Journal Name:
Mathematics of Computation
ISSN:
0025-5718
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. We show that if L 1 \mathcal {L}_1 and L 2 \mathcal {L}_2 are linear transformations from Z d \mathbb {Z}^d to Z d \mathbb {Z}^d satisfying certain mild conditions, then, for any finite subset A A of Z d \mathbb {Z}^d , | L 1 A + L 2 A | ≥<#comment/> ( | det ( L 1 ) | 1 / d + | det ( L 2 ) | 1 / d ) d | A | −<#comment/> o ( | A | ) . \begin{equation*} |\mathcal {L}_1 A+\mathcal {L}_2 A|\geq \left ( |\det (\mathcal {L}_1)|^{1/d}+|\det (\mathcal {L}_2)|^{1/d} \right )^d|A|- o(|A|). \end{equation*} This result corrects and confirms the two-summand case of a conjecture of Bukh and is best possible up to the lower-order term for certain choices of L 1 \mathcal {L}_1 and L 2 \mathcal {L}_2 . As an application, we prove a lower bound for | A + λ<#comment/> ⋅<#comment/> A | |A + \lambda \cdot A| when A A is a finite set of real numbers and λ<#comment/> \lambda is an algebraic number. In particular, when λ<#comment/> \lambda is of the form ( p / q ) 1 / d (p/q)^{1/d} for some p , q , d ∈<#comment/> N p, q, d \in \mathbb {N} , each taken as small as possible for such a representation, we show that | A + λ<#comment/> ⋅<#comment/> A | ≥<#comment/> ( p 1 / d + q 1 / d ) d | A | −<#comment/> o ( | A | ) . \begin{equation*} |A + \lambda \cdot A| \geq (p^{1/d} + q^{1/d})^d |A| - o(|A|). \end{equation*} This is again best possible up to the lower-order term and extends a recent result of Krachun and Petrov which treated the case λ<#comment/> = 2 \lambda = \sqrt {2}
    more » « less
  2. We show that for primes N , p ≥<#comment/> 5 N, p \geq 5 with N ≡<#comment/> −<#comment/> 1 mod p N \equiv -1 \bmod p , the class number of Q ( N 1 / p ) \mathbb {Q}(N^{1/p}) is divisible by p p . Our methods are via congruences between Eisenstein series and cusp forms. In particular, we show that when N ≡<#comment/> −<#comment/> 1 mod p N \equiv -1 \bmod p , there is always a cusp form of weight 2 2 and level Γ<#comment/> 0 ( N 2 ) \Gamma _0(N^2) whose ℓ<#comment/> \ell th Fourier coefficient is congruent to ℓ<#comment/> + 1 \ell + 1 modulo a prime above p p , for all primes ℓ<#comment/> \ell . We use the Galois representation of such a cusp form to explicitly construct an unramified degree- p p extension of Q ( N 1 / p ) \mathbb {Q}(N^{1/p})
    more » « less
  3. If I I is an ideal in a Gorenstein ring S S , and S / I S/I is Cohen-Macaulay, then the same is true for any linked ideal I I’ ; but such statements hold for residual intersections of higher codimension only under restrictive hypotheses, not satisfied even by ideals as simple as the ideal L n L_{n} of minors of a generic 2 ×<#comment/> n 2 \times n matrix when n > 3 n>3 . In this paper we initiate the study of a different sort of Cohen-Macaulay property that holds for certain general residual intersections of the maximal (interesting) codimension, one less than the analytic spread of I I . For example, suppose that K K is the residual intersection of L n L_{n} by 2 n −<#comment/> 4 2n-4 general quadratic forms in L n L_{n} . In this situation we analyze S / K S/K and show that I n −<#comment/> 3 ( S / K ) I^{n-3}(S/K) is a self-dual maximal Cohen-Macaulay S / K S/K -module with linear free resolution over S S . The technical heart of the paper is a result about ideals of analytic spread 1 whose high powers are linearly presented. 
    more » « less
  4. We formulate and prove a Conner–Floyd isomorphism for the algebraic K-theory of arbitrary qcqs derived schemes. To that end, we study a stable ∞<#comment/> \infty -category of non- A 1 \mathbb {A}^1 -invariant motivic spectra, which turns out to be equivalent to the ∞<#comment/> \infty -category of fundamental motivic spectra satisfying elementary blowup excision, previously introduced by the first and third authors. We prove that this ∞<#comment/> \infty -category satisfies P 1 \mathbb {P}^1 -homotopy invariance and weighted A 1 \mathbb {A}^1 -homotopy invariance, which we use in place of A 1 \mathbb {A}^1 -homotopy invariance to obtain analogues of several key results from A 1 \mathbb {A}^1 -homotopy theory. These allow us in particular to define a universal oriented motivic E ∞<#comment/> \mathbb {E}_\infty -ring spectrum M G L \mathrm {MGL} . We then prove that the algebraic K-theory of a qcqs derived scheme X X can be recovered from its M G L \mathrm {MGL} -cohomology via a Conner–Floyd isomorphism\[ M G L ∗<#comment/> ∗<#comment/> ( X ) ⊗<#comment/> L Z [ β<#comment/> ±<#comment/> 1 ] ≃<#comment/> K ∗<#comment/> ∗<#comment/> ( X ) , \mathrm {MGL}^{**}(X)\otimes _{\mathrm {L}{}}\mathbb {Z}[\beta ^{\pm 1}]\simeq \mathrm {K}{}^{**}(X), \]where L \mathrm {L}{} is the Lazard ring and K p , q ( X ) = K 2 q −<#comment/> p ( X ) \mathrm {K}{}^{p,q}(X)=\mathrm {K}{}_{2q-p}(X) . Finally, we prove a Snaith theorem for the periodized version of M G L \mathrm {MGL}
    more » « less
  5. Let ( G / Γ<#comment/> , R a ) (G/\Gamma ,R_a) be an ergodic k k -step nilsystem for k ≥<#comment/> 2 k\geq 2 . We adapt an argument of Parry [Topology 9 (1970), pp. 217–224] to show that L 2 ( G / Γ<#comment/> ) L^2(G/\Gamma ) decomposes as a sum of a subspace with discrete spectrum and a subspace of Lebesgue spectrum with infinite multiplicity. In particular, we generalize a result previously established by Host–Kra–Maass [J. Anal. Math.124(2014), pp. 261–295] for 2 2 -step nilsystems and a result by Stepin [Uspehi Mat. Nauk24(1969), pp. 241–242] for nilsystems G / Γ<#comment/> G/\Gamma with connected, simply connected G G
    more » « less