{"id":"97dd2a40-3ff1-4804-9e50-d3ab18a6781a","arxiv_id":"2501.19130","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":2.0,"correctness_risk":"low","formal_verification":"none","parameter_count":5,"one_line_summary":"HMC with exact Fourier acceleration is exact for quadratic actions and only mildly worse for perturbed ones; for generic targets the guide recommends random long trajectories, radial updates, and a regularized kinetic term.","lead":"This paper is a practical guide to hybrid Monte Carlo, a sampling algorithm used across physics and statistics, with recent upgrades that make it exact for Gaussian problems. It collects the full recipe including Fourier acceleration and radial updates, and shows where the method still struggles.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified.","rationale":"The paper is a proceedings-style practical guide whose central claims are delegated to Ref. [2]. Re-deriving the EFA update: with H = 1/2 p^T M^{-1} p + 1/2 x^T M x, the normal-mode equations are y'' = -y and q'' = -q, and after time T=pi/2 the position maps to y = q0/omega^2, i.e. x_f = M^{-1} p0. Since p0 is sampled as sqrt(M) r with r ~ N(0,1), x_f ~ N(0, M^{-1}) and is independent of x0. The leap-frog composition of EFA half-steps exactly realizes this rotation when V=0, so the quadratic-action equivalence is exact. For perturbative V, the leading correction is O(epsilon) in the trajectory and O(epsilon^2) in Delta H, consistent with high acceptance and near-minimal autocorrelation; the paper's numerical examples, while lacking error bars, support this. The reader's flagged assumption that M must be known and diagonalizable is a genuine scope limitation, especially for non-translationally-invariant systems, but it is explicitly acknowledged in the text and does not undermine the mathematical content of the central claim. Accordingly, the verdict should remain ACCEPT.","tokens_in":6238,"tokens_out":13105,"duration_ms":148450,"concrete_test":"Verify Algorithm 1 on a 32-dimensional Gaussian with a non-diagonal positive-definite M, using one EFA trajectory per sample and T=pi/2: collect 10^5 outputs, record the acceptance rate (should be 1 within Monte Carlo error), and run a two-sample goodness-of-fit test against direct Cholesky samples from N(0, M^{-1}) to confirm the output distribution matches the target.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is sound for its stated regime: for a positive-definite quadratic action, Algorithm 3 exactly integrates the harmonic dynamics, and with T=pi/2 the output x_f = M^{-1} p is independent of x0, so every trajectory produces an exact target sample with acceptance 1. The 'only perturbatively worse' and 'viable for arbitrary actions' parts are explicitly qualified in Sec. 3.1.4, which concedes that in the strongly anharmonic case no optimal kinetic term can be derived without further knowledge of V(x). The main practical limitation, also noted by the reader, is that EFA requires the eigen-decomposition of M; this affects scope and cost, not the mathematical correctness of the claim. No load-bearing correctness concern was found.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper is a compact procedural guide to hybrid Monte Carlo with exact Fourier acceleration (EFA). It states that for actions S(x)=½x^T M x + V(x) with V(x) small, HMC with EFA solves the harmonic part exactly in the eigenbasis of M (Algorithms 1–3), and that with trajectory length T=π/2 each trajectory produces a sample from the quadratic target with acceptance probability one. It then collects adjustments needed for generic actions: randomising trajectory lengths, choosing long trajectories, including radial updates on non-compact spaces, and regularising the kinetic term for strongly anharmonic actions. Numerical illustrations are given for the SSH model, the 2D Ising model, and pure gauge theory, and the paper closes with a summary of the recommended practical recipe.","tokens_in":6394,"tokens_out":6442,"duration_ms":66343,"significance":"If the central claims hold, the recipe gives a concrete and simple way to achieve near-optimal sampling for near-Gaussian high-dimensional targets, which is directly relevant for lattice field theory and condensed-matter simulations. The algorithmic core is correct, and the self-contained presentation of Algorithms 1–4, together with the pointer to open-source code in Ref. [2], is a useful service to practitioners. The main caveats are that the 'only perturbatively worse' statement is not derived here but is delegated to Ref. [2], that the efficiency claim for arbitrary actions is explicitly limited in Sec. 3.1.4, and that the numerical evidence is presented without error bars or full parameter reporting. These caveats affect the strength of the advertised claims but not the correctness of the quadratic-action equivalence.","major_comments":[],"minor_comments":[{"comment":"The phrase 'only perturbatively worse for perturbative deviations' is stronger than anything demonstrated in this manuscript; since all derivations are deliberately delegated to Ref. [2], please either state the precise perturbation result with its assumptions or soften the wording to 'as shown in Ref. [2]'.","section":"Abstract and Sec. 2"},{"comment":"The dashed 'prediction' lines are normalized to the measured autocorrelation time at T=π/2, so they test only the functional scaling, not the absolute value of the autocorrelation time; please label them as scaling predictions rather than parameter-free predictions.","section":"Figs. 1 and 2"},{"comment":"The integrated autocorrelation times are reported without error bars, and the integrator step size h is not specified, so the statistical significance of the differences between EFA and no-FA, and of the agreement with the dashed lines, cannot be assessed; please add error bars (for example using the method of Ref. [9]) and report h and acceptance rates precisely.","section":"Figs. 1–3"},{"comment":"The algorithm assumes M is positive definite with eigenvalues ω_i^2>0; for clarity, please state this assumption explicitly and indicate what should be done if M has zero or negative modes.","section":"Algorithm 3"},{"comment":"Please clarify that the radial update multiplies the whole vector by e^γ and that the acceptance probability follows from changing variables in R^d; the current description is correct but could be clearer for readers who are not already familiar with radial updates.","section":"Algorithm 4"},{"comment":"The left panel uses physical parameters with ω0≈6 meV while the right panel uses ω0=1 eV; please add a sentence explaining this difference in units so that the reader does not mistake the two panels for the same physical regime.","section":"Sec. 2.1"},{"comment":"Reference [12] lists the No-U-Turn sampler as 'Homan and Gelman'; the correct spelling is 'Hoffman and Gelman'.","section":"References"},{"comment":"The first sentence reads 'The authors thanks'; this should be 'The author thanks'.","section":"Acknowledgements"}],"recommendation":"minor_revision","confidential_remarks":"This is an extended proceedings note whose technical content is largely delegated to the author's own Ref. [2]. That delegation is acceptable for a guide, but the abstract should not overstate what is proved in the paper itself. The refereed central claim—that for quadratic actions EFA-HMC with T=π/2 samples directly from N(0,M^-1)—is correct and self-contained in Algorithms 1–3. The main weakness is the numerical support: error bars and step-size information are missing, and the 'prediction' lines are normalized to data. These are fixable in revision and do not undermine the algorithmic derivation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a practical guide, not a research paper, and once you know that it is a good one. The core trick—exact integration of the harmonic part of the action, making HMC exact for Gaussian targets—was already in the author's own Ref. [2]. What this paper adds is a compact, self-contained recipe with explicit algorithms and numerical illustrations. That is worth having.\n\nThe algorithms are correct. For a quadratic action, Algorithm 3 exactly solves the equations of motion, so with trajectory length T=pi/2 the proposed configuration is independent of the starting point and acceptance is 1. The guide is also honest about its limits: Sec. 3.1.4 admits that for strongly anharmonic actions there is no way to derive an optimal kinetic term without more information about the potential, and Sec. 3.2 says the gauge theory results are inconclusive. That is the right level of candor.\n\nSoft spots: the prediction lines in Figs. 1 and 2 are normalized to measured autocorrelation values, so they are partly fits and not clean tests of the theory. The figures have no error bars, which makes it hard to judge how much weight to put on the comparisons. Baselines like the no-FA trajectory lengths are hand-tuned, which is fine for illustration but not a systematic benchmark. Also, the guide assumes you have a usable eigendecomposition of the harmonic matrix M; for translationally invariant problems that is a Fourier transform, but for generic problems it can be as expensive as the sampling itself. The self-citation pattern is heavy, but here it is natural: the author is summarizing his own line of work, and the derivations are explicitly delegated to Ref. [2].\n\nWho is this for? Physicists or computational scientists who want to implement EFA with radial updates without wading through the longer papers. It is a useful 'how to' reference for a lattice field theory group. It will not move the research frontier, but it does not pretend to.\n\nMy recommendation: send it to peer review. It is a solid proceedings-style paper that deserves a careful referee. The referee should ask for error bars or at least an explicit statement that they are omitted, and should double-check the normalization of the prediction curves. Nobody should desk-reject it just because it is not a new scientific result.","headline":"A clear, honest recipe for HMC with exact Fourier acceleration; no new science, but the algorithms are correct and the guide is genuinely useful.","tokens_in":6874,"tokens_out":1518,"would_cite":true,"duration_ms":16646,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["02.70.Uu","11.15.Ha"],"model":"deepseek-v4-flash","headline":"The paper claims that hybrid Monte Carlo with exact Fourier acceleration becomes equivalent to direct sampling for quadratic actions, and remains viable for generic actions when augmented with long randomised trajectories and radial…","keywords":["hybrid Monte Carlo","exact Fourier acceleration","autocorrelation time","lattice field theory","radial updates","Hamiltonian Monte Carlo","Gaussian sampling","MCMC"],"falsifier":"Simulate a translationally invariant Gaussian target on a $32\\times32$ lattice using EFA and measure the integrated autocorrelation time $\\tau_{\\mathrm{int}}$ of a nonlinear observable such as $\\langle x^4\\rangle$; if $\\tau_{\\mathrm{int}}$ grows with lattice size instead of staying near its minimum of 0.5, the equivalence to direct sampling fails. Alternatively, for a strongly anharmonic action, compare EFA with plain HMC at equal compute time; if EFA's acceptance or effective sample size is not better, the claimed advantage has not materialised.","tokens_in":6038,"feed_emoji":"🎲","tokens_out":8661,"duration_ms":78186,"temperature":0.7,"pith_summary":"The paper aims to establish a practical recipe for hybrid Monte Carlo (HMC) sampling that reaches near-optimal efficiency for a wide class of probability distributions. Its central claim is that coupling HMC with exact Fourier acceleration (EFA) makes the sampler equivalent to direct sampling for quadratic actions $S(x)=\\frac12 x^\\mathsf{T} M x$, i.e. normal distributions, and only perturbatively worse when a small anharmonic perturbation $V(x)$ is added. The author argues that EFA, together with randomised sufficiently long trajectories and radial updates on non-compact spaces, keeps the algorithm viable for arbitrary continuous distributions. The argument is supported by the analytical results of the companion reference and by numerical benchmarks on the SSH model, the Ising model, and lattice gauge theories, all of which show integrated autocorrelation times near the theoretical floor.","feed_headline":"Fourier acceleration makes HMC match direct sampling on Gaussians","feed_subtitle":"A hands-on recipe keeps autocorrelation near its floor and acceptance near one for HMC sampling.","key_machinery":"The central object is the split Hamiltonian $H=\\frac12 p^\\mathsf{T} M^{-1} p + S(x)$, with the exact Fourier acceleration (EFA) step of Algorithm 3 as the load-bearing mechanism. EFA uses the eigendecomposition $M=\\Omega\\, \\mathrm{diag}(\\omega^2)\\, \\Omega^\\dagger$ (often a Fourier transform for translationally invariant systems) to evolve harmonic eigenmodes exactly: after rotating to the eigenbasis, each mode is advanced by the exact rotation $y_i(h)=\\cos(h)\\, y0_i + (1/\\omega_i^2)\\, \\sin(h)\\, q0_i$ and $q_i(h)=\\cos(h)\\, q0_i - \\omega_i^2\\, \\sin(h)\\, y0_i$. This eliminates the sampling error coming from the quadratic part, leaving only the anharmonic perturbation to be handled by the integrator; the leap-frog step and the Metropolis acceptance rule then convert the near-conservation of $H$ into a high acceptance rate and minimal autocorrelation.","core_discovery":"For a distribution $P(x)\\propto \\mathrm{e}^{-S(x)}$ with harmonically dominated action $S=\\frac12 x^\\mathsf{T} M x + V(x)$, the paper's claim is that HMC with EFA samples the harmonic part exactly: the molecular dynamics solves the equation of motion for each eigenmode of $M$ with a closed cosine/sine step, so $\\Delta H$ comes only from the anharmonic perturbation. With trajectory length $T=\\pi/2$ and acceptance near one, the integrated autocorrelation time stays close to its minimum and the algorithm behaves like direct sampling. The claim extends perturbatively, so small deviations from the Gaussian case are only mildly worse, and it remains a viable sampler for arbitrary actions when three adjustments are made: randomised trajectory length, sufficiently long trajectories, and radial updates for non-compact spaces. Algorithms 1–4 implement this as a turn-key numerical recipe.","pith_inferences":["For problems without translational symmetry, the eigendecomposition of $M$ is a genuine one-time cost that the paper does not budget for; the method's advantage is most certain when $M$ is diagonalisable by a fast transform.","A quantitative follow-up would be an analytic bound on acceptance-rate or autocorrelation loss as a function of the anharmonic fraction $|V(x)|/(x^\\mathsf{T} M x)$, giving users a threshold for when EFA is worth the extra implementation effort.","Folding EFA into adaptive HMC samplers that already randomise path length (such as the No-U-Turn sampler) could yield a fully automatic near-optimal sampler for generic actions, though the paper stops short of testing that combination."],"forward_implications":["With EFA, users of HMC on Gaussian and near-Gaussian targets can set the trajectory length to $T=\\pi/2$ and use a single leap-frog step, obtaining acceptance rates that stay in the 65–80% range and integrated autocorrelation times below about 0.7.","For translationally invariant systems, the eigendecomposition of $M$ is a Fourier transform, so EFA adds almost no overhead and applies directly to lattice problems with minimal extra cost.","Randomising the trajectory length and choosing it long enough on average prevents the slow-decorrelation modes that appear when the eigenfrequencies of the HMC dynamics are unknown.","On non-compact spaces, radial updates restore exponential convergence and can overcome barrier-induced ergodicity problems, completing the recipe for generic actions."],"supporting_citations":[{"why":"Introduces the hybrid Monte Carlo algorithm that the guide optimises; the base sampler being accelerated.","marker":"[1]"},{"why":"Derives exact Fourier acceleration and the minimal-autocorrelation predictions (corollaries 1 and 2) that ground the equivalence claim.","marker":"[2]"},{"why":"Original Fourier acceleration for lattice field theories; provides the idea of diagonalising the harmonic part via Fourier transform.","marker":"[5]"},{"why":"Shows HMC with long and random trajectories is efficient for Gaussian sampling, supporting the trajectory-length advice for generic actions.","marker":"[11]"},{"why":"Establishes convergence properties of HMC on non-compact manifolds, motivating the radial-update correction.","marker":"[15]"},{"why":"Introduces radial updates with exponential speed-up; the paper uses this as Algorithm 4 to restore ergodicity on non-compact spaces.","marker":"[16]"},{"why":"Supplies the SSH-model simulation benchmark showing EFA's near-minimal autocorrelation at physical parameters.","marker":"[8]"},{"why":"Provides the Ising-model HMC benchmark used to demonstrate the effect of trajectory length and the onset of critical slowing down.","marker":"[13]"}],"fun_headline_variants":["Fourier-accelerated HMC samples Gaussians exactly","HMC with EFA: exact Gaussian sampling","Recipe: Fourier-accelerated HMC for exact Gaussian sampling","Fourier acceleration makes HMC exact on Gaussians","HMC + EFA equals direct sampling on Gaussians"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The recipe's efficiency rests on knowing the harmonic matrix $M$ and being able to diagonalise it (ideally by a Fourier transform), and on the harmonic part dominating the action so that the residual $V(x)$ is a small perturbation.","fun_headline_variants_meta":{"raw":{"variants":["Fourier-accelerated HMC samples Gaussians exactly","HMC with EFA: exact Gaussian sampling","Recipe: Fourier-accelerated HMC for exact Gaussian sampling","Fourier acceleration makes HMC exact on Gaussians","HMC + EFA equals direct sampling on Gaussians"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001541,"raw_usage":{"total_tokens":6102,"prompt_tokens":818,"completion_tokens":5284,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":434,"completion_tokens_details":{"reasoning_tokens":5203}},"tokens_in":434,"tokens_out":5284,"duration_ms":34829,"temperature":1.0,"reasoning_tokens":5203,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T21:12:04.447120+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate a translationally invariant Gaussian target on a $32\\times32$ lattice using EFA and measure the integrated autocorrelation time $\\tau_{\\mathrm{int}}$ of a nonlinear observable such as $\\langle x^4\\rangle$; if $\\tau_{\\mathrm{int}}$ grows with lattice size instead of staying near its minimum of 0.5, the equivalence to direct sampling fails. Alternatively, for a strongly anharmonic action, compare EFA with plain HMC at equal compute time; if EFA's acceptance or effective sample size is not better, the claimed advantage has not materialised.","supporting_citations":[{"cited_title":"Batrouni, G.R","cited_arxiv_id":null,"evidence_quote":"Original Fourier acceleration for lattice field theories; provides the idea of diagonalising the harmonic part via Fourier transform."},{"cited_title":"Apers, S","cited_arxiv_id":null,"evidence_quote":"Shows HMC with long and random trajectories is efficient for Gaussian sampling, supporting the trajectory-length advice for generic actions."},{"cited_title":"Kennedy and X","cited_arxiv_id":null,"evidence_quote":"Establishes convergence properties of HMC on non-compact manifolds, motivating the radial-update correction."},{"cited_title":"Exponential speed up in Monte Carlo sampling through Radial Updates","cited_arxiv_id":"2411.18218","evidence_quote":"Introduces radial updates with exponential speed-up; the paper uses this as Algorithm 4 to restore ergodicity on non-compact spaces."},{"cited_title":"First-principle quantum Monte-Carlo study of charge carrier mobility in organic molecular semiconductors","cited_arxiv_id":"2312.14914","evidence_quote":"Supplies the SSH-model simulation benchmark showing EFA's near-minimal autocorrelation at physical parameters."},{"cited_title":"Ostmeyer, E","cited_arxiv_id":null,"evidence_quote":"Provides the Ising-model HMC benchmark used to demonstrate the effect of trajectory length and the onset of critical slowing down."}],"review_version":1}