{"id":"cf0b3ae4-c3cb-4f4d-90cb-c1beb11c1a52","arxiv_id":"2501.01396","paper_version":3,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"A discrete-lattice version of the Johnson-Lindenstrauss lemma is attempted, but a scaling error in the lower distortion bound leaves the central claim unproved.","lead":"A short note proposes a Johnson-Lindenstrauss style dimension reduction for point sets in discrete lattices, embedding them into a lower-dimensional lattice while preserving distances up to a small distortion. The proof combines the JL lemma with Ziegler's rotation theorem, but contains scaling inconsistencies that break the stated claim.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof constructs a map on λD, not on D; (2.3) is off by a factor λ from (1.1), so the claimed distortion for F:D→(1/λ0)Z^k is not established.","rationale":"The paper's goal is a discrete-grid Johnson–Lindenstrauss-type embedding for point sets in (λ/λ0)Z^d. The proof outline—JL projection, Ziegler-style rotation/rounding, then a distortion estimate—is plausible, but the decisive step is where the constructed map is transferred back to the original domain D. The proof explicitly defines F on λx_i, and equation (2.3) bounds the output distance by a quantity proportional to ‖λx_i−λx_j‖, while (1.1) requires the bound against ‖x_i−x_j‖. This is not a harmless typo: rescaling z_i by λ to fix the upper bound destroys the lower bound by the same factor. The reader's weakest_assumption identified exactly this scaling mismatch, and the numerical test isolates the factor-λ gap. Because the proof of the central proposition fails at this point and no other argument in the manuscript fills the gap, the paper should not be accepted as establishing the stated result. Since the identified concern is the same as the reader's, no change to the reader's REJECT verdict is needed.","tokens_in":3941,"tokens_out":10842,"duration_ms":106870,"concrete_test":"Set d=2, λ0=1, ε=0.1, λ=100, and D={0,100e1}. Verify whether the proof's construction can satisfy (1.1). If F:D→Z^k exists as claimed, then for x=0 and y=100e1 we must have ‖F(x)−F(y)‖ ≤ (1.1+0.001)·100 ≈ 110.01. But (2.3), applied to the z_i actually constructed by rounding λΦ(x) and λΦ(y), gives only ‖z1−z2‖ ≤ ε + λ(1+ε)‖x−y‖ ≈ 11000. The gap is exactly the missing factor λ, so the chain (2.3)–(2.4) cannot produce the stated upper bound.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The Main Proposition requires a map F:D→(1/λ0)Z^k satisfying (1.1) for x,y∈D. In the proof, however, the map is introduced as 'F: λx_i ↦ (1/λ0)z_i', where z_i are integer vectors obtained by rounding ρ(λΦ(x_i)) to (1/λ0)Z^k. Thus the proof actually controls an embedding of the dilated set λD, not of D. Equation (2.3) gives (1/λ0)‖z_i−z_j‖ ≤ (1+ε+ε/(λλ0))‖λx_i−λx_j‖, which is a factor λ larger than the right-hand side of (1.1) when x_i,x_j are viewed as points of D. Equation (2.4) is stated relative to ‖x_i−x_j‖, so the two inequalities are normalized at incompatible scales. If one rescales z_i to force the upper bound to match (1.1), the target grid changes to 1/(λλ0)Z^k and the lower bound (2.4) acquires an extra factor 1/λ. Consequently (2.3) and (2.4) do not imply (1.1); the central proposition is unsupported. A related scale error appears earlier when Dflat is described as lying in (1/λ0)√kZ^k, although D⊂(λ/λ0)Z^d forces an additional factor λ in the lattice spacing of Φ(D).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a discrete version of the Johnson-Lindenstrauss lemma: for point sets lying in the lattice (λ/λ0)Z^d and in a ball of radius λN0, it claims the existence of an embedding into (1/λ0)Z^k with dimension k = O(ε^{-2} log d) and distortion close to 1. The proof combines the classical JL lemma with a theorem of T. Ziegler on rotating dilated finite sets near the integer lattice, together with a rounding step into (1/λ0)Z^k. The central assertion is Proposition 1.1, and the proof is contained in Section 2.","tokens_in":4223,"tokens_out":7899,"duration_ms":72180,"significance":"If the main proposition were correct, it would provide a natural discrete counterpart to JL dimensionality reduction, with the target space constrained to a lattice of the same spacing as the input lattice. The question is interesting and the chosen tools (JL, Ziegler's theorem, uniform distribution) are relevant. However, the proof as written contains a fundamental scaling error: the map is constructed on the dilated set λD while the theorem requires a map on D. The result may still be salvageable with a corrected target lattice or a reformulated distortion bound, but the current manuscript does not establish the stated claim. There are no machine-checked proofs or reproducible code in the paper.","major_comments":[{"comment":"The map F is defined as F: λx_i ↦ (1/λ0) z_i, i.e., on the dilated set λD, whereas Proposition 1.1 requires F: D → (1/λ0)Z^k. Consequently, equation (2.3) bounds (1/λ0)‖z_i−z_j‖ in terms of ‖λx_i−λx_j‖, which is λ times the norm ‖x_i−x_j‖ appearing in (1.1). Equation (2.4) provides a lower bound relative to ‖x_i−x_j‖, so the two inequalities are normalized at incompatible scales. Writing G(x) = F(λx) for x∈D, equations (2.3)–(2.4) yield λ(1−ε−ε/(λλ0))‖x_i−x_j‖ ≤ ‖G(x_i)−G(x_j)‖ ≤ λ(1+ε+ε/(λλ0))‖x_i−x_j‖ (up to additive ε terms), not the constant-factor distortion claimed in (1.1). The proof therefore does not establish the proposition as stated.","section":"§2, above (2.3)"},{"comment":"The claim that Dflat is a subset of (1/λ0)√k Z^k, based on Φ(x) = (1/√k) R x^T with R an {0,1}-valued matrix, omits the factor λ: since x_i ∈ (λ/λ0)Z^d, the image coordinates lie in (λ/(λ0√k))Z, not (1/(λ0√k))Z. This is the same scaling error in a different location and further affects the subsequent application of Ziegler's theorem and the rounding step.","section":"§2, paragraph on Dflat"},{"comment":"The lemma states that for every λ ≥ λ1 there is a rotation ρ with d(ρ(λD), Z^k) < ε for all D ⊂ tZ^k ∩ B_N. The proof, however, only produces a single integer n1 such that d(n1D, Z^k) < ε, and then says the rest follows as in [2]. No argument is given for the uniformity in λ ≥ λ1, and the proof does not establish the lemma as stated. Since Lemma 2.1 is a load-bearing ingredient in the proof of the main proposition, this is a significant gap.","section":"Lemma 2.1"},{"comment":"The proof asserts, by Ziegler's Theorem and Lemma 2, that there exists λ1 depending only on ε,k,N0 such that for all λ ≥ λ1 there is a rotation ρ with d(ρ(λDflat), (1/λ0)Z^k) ≤ ε/λ0. Ziegler's Theorem provides large integer multiples of a fixed finite set close to the lattice; here the set λ0Dflat depends on λ through the factor λ in the lattice spacing of Dflat. The threshold λ1 therefore may need to depend on Dflat (i.e., on the particular point set D), but the proposition requires a λ1 that works uniformly for all λ ≥ λ1 for each fixed D. This uniformity is not justified, and no proof is supplied.","section":"§2, use of Ziegler's Theorem"}],"minor_comments":[{"comment":"The abstract contains typos such as 'Johnson-Lindenstra uss flattening' and the phrase 'suitably chosen' for ε is vague; the hypotheses on ε and on the parameters deserve precise statements.","section":"Abstract"},{"comment":"The final inequality in (2.3) appears to require that ε ≤ (ε/λ0)‖x_i−x_j‖, which uses the minimum spacing of points in D. This is not stated or proved; it may hold for sufficiently large λ, but the justification is missing.","section":"§2, equation (2.3)"},{"comment":"The phrase 'since a translation by a vector is an isometry, we can assume the origin of R^k is the circumcenter of Dflat' requires justification that the translation preserves the lattice property of the point set; not all translations preserve the membership in (λ/(λ0√k))Z^k.","section":"§2, "},{"comment":"Reference [5] appears garbled ('M. K/suppress lopotek'); this should be corrected to the actual author name.","section":"References"}],"recommendation":"reject","confidential_remarks":"The central theorem is not established because of a fundamental scaling error in the definition of the map and in the comparison of distances. The error is not a minor presentation issue, and it is unlikely to be fixable without changing the statement of the main proposition (e.g., to a target lattice (λ/λ0)Z^k or to a distortion bound of order λ). The paper also has an incomplete proof of a key lemma. Given these issues, I recommend rejection. I would be willing to look at a revised version if the authors provide a corrected proof of a properly stated theorem."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The main take: the paper's idea is new, but the proof as written does not establish the stated result. The construction works on a scaled copy of the point set, and the upper and lower bounds are normalized at incompatible scales. That is a load-bearing error, not a typo.\n\nWhat's actually new: Viera proposes a discrete-subgroup version of the JL lemma, targeting a lattice (1/λ0)Z^k, and brings two tools to bear: Ziegler's rotation theorem and uniform distribution modulo 1. The explicit error term ε/(λλ0) and the use of a large scaling λ are a reasonable way to get integer-valued embeddings. The combination is not in the cited literature, and the author does not oversell it as a new framework.\n\nThe soft spot: the proof of Proposition 1.1 defines F on λD, not D. Equation (2.3) bounds (1/λ0)‖z_i−z_j‖ by (1+ε+ε/(λλ0))‖λx_i−λx_j‖, which is a factor λ larger than the upper bound required in (1.1). Equation (2.4) is stated relative to ‖x_i−x_j‖. So the two inequalities cannot be chained to give (1.1). If you rescale z_i to force the upper bound to match (1.1), the target lattice becomes (1/(λλ0))Z^k and the lower bound picks up an extra 1/λ. The main claim is therefore unsupported. There are also smaller issues: Lemma 2.1 is deferred to [2] for the hard part, the reference to a {0,1}-matrix JL transform is not standard (usually ±1), and the claim that Dflat lies in (1/λ0)√k Z^k appears to be off by a factor λ.\n\nOn the plus side, the paper is short and honest; it cites the right literature and does not hide its reliance on Ziegler and uniform distribution. It is just the central chain that fails.\n\nIf you are deciding whether to referee it: I would say it deserves a look, but with the expectation that the author needs to either fix the scaling, which may be possible by redefining the target lattice, or substantially revise the statement. As it stands, I would not accept the proof. For a reading group, it is a reasonable example of how scaling errors creep into JL-style proofs. I would not cite it yet.","headline":"A genuinely new combination of ideas, but the proof's scaling error leaves the main proposition unproven.","tokens_in":4770,"tokens_out":5177,"would_cite":false,"duration_ms":47123,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["46B85"],"pacs":[],"model":"deepseek-v4-flash","headline":"Discrete subgroups admit JL flattening into $O(\\varepsilon^{-2}\\log d)$ dimensions.","keywords":["Johnson–Lindenstrauss lemma","discrete subgroup","lattice embedding","dimensionality reduction","metric distortion","uniform distribution modulo one","Ziegler's theorem","quantized data"],"falsifier":"For $\\lambda_0=2$ and $\\varepsilon=1/4$, take two points in $D$ at the minimal lattice separation $\\lambda/\\lambda_0$ apart. Substituting the unscaled distance $\\|x_i-x_j\\|=1/\\lambda_0$ into the proof's inequality (2.3) reduces its last step to the requirement $\\varepsilon\\le \\varepsilon/\\lambda_0^2$, which is false for $\\lambda_0>1$; checking (1.1) for this pair therefore settles whether the claimed uniform distortion holds for all pairs.","tokens_in":3692,"feed_emoji":"📐","tokens_out":13397,"duration_ms":110043,"temperature":0.7,"pith_summary":"The paper claims a discrete version of the Johnson–Lindenstrauss lemma for point sets that live in a fine lattice $\\frac{\\lambda}{\\lambda_0}\\mathbb{Z}^d$ inside a ball of radius $\\lambda N_0$. After the lattice is scaled by a sufficiently large integer $\\lambda$, any $d$ such points can be mapped into the coarser grid $\\frac{1}{\\lambda_0}\\mathbb{Z}^k$, where $k=O(\\varepsilon^{-2}\\log d)$, with pairwise distances preserved up to a factor $1+\\varepsilon+\\frac{\\varepsilon}{\\lambda\\lambda_0}$. The result matters because realistic high-dimensional data is quantised rather than continuous, and this says the classical flattening can be adapted to respect the grid structure while keeping both the number of decimals and the magnitude of the output bounded.","feed_headline":"Discrete lattice data flattens to ~log d dimensions","feed_subtitle":"A discrete analogue of the Johnson–Lindenstrauss lemma preserves distances up to 1+epsilon on a coarse grid.","key_machinery":"The argument is carried by three ingredients. The Johnson–Lindenstrauss lemma provides a linear map $\\Phi:\\mathbb{R}^d\\to\\mathbb{R}^k$ that is an $(1+\\varepsilon)$-embedding on the input points. Ziegler's theorem supplies, for each sufficiently large dilation $l$, a rotation $\\rho\\in SO(k)$ such that the dilates of the projected points are within $\\varepsilon$ of $\\mathbb{Z}^k$; this is the step that forces the large scale $\\lambda$. A one-dimensional uniform-distribution lemma for irrational $t$ guarantees that dilations of $t\\mathbb{Z}$ modulo $1$ are dense, which is used to find the admissible scale $\\lambda_1$. The final map $F$ sends a point $\\lambda x_i$ to the nearest point of $\\frac{1}{\\lambda_0}\\mathbb{Z}^k$ after the rotation $\\rho$ is applied.","core_discovery":"The central discovery is that a Johnson–Lindenstrauss projection can be composed with a rotation and a rounding step so that the output lies exactly on a prescribed grid, with the quantisation error absorbed by a large dilation of the input lattice. Concretely, the author proves that for every $d,\\lambda_0,N_0\\in\\mathbb{N}$ and $\\varepsilon\\in(0,\\frac{1}{\\lambda_0+1})$ there is a dimension $k=O(\\varepsilon^{-2}\\log d)$ such that, once $\\lambda$ is taken large enough, any $d$-point set in $\\frac{\\lambda}{\\lambda_0}\\mathbb{Z}^d\\cap B(0,\\lambda N_0)$ admits a map $F$ into $\\frac{1}{\\lambda_0}\\mathbb{Z}^k$ satisfying the two-sided distortion bound for all pairs. The construction uses the classical JL lemma, Ziegler's theorem on rotating dilated finite sets close to the integer lattice, and a uniform-distribution lemma to justify the existence of the dilation scale.","pith_inferences":["If the scaling gap in the proof can be closed, a constructive version with explicit $\\lambda_1$ would give a fully discrete JL transform usable in practice on quantised data.","The dependence on Ziegler's theorem suggests the embedding is non-constructive; obtaining it algorithmically may require an effective uniform-distribution statement.","A natural test is whether the theorem survives when $\\varepsilon$ is allowed to be larger than $1/(\\lambda_0+1)$; the current restriction couples distortion to grid coarseness.","The argument may extend to other discrete subgroups, such as weighted integer lattices, provided an analogue of Ziegler's rotation theorem holds there."],"forward_implications":["If the main proposition is correct, any finite data set confined to a sufficiently fine lattice can be embedded into a coarser lattice of dimension $O(\\varepsilon^{-2}\\log d)$ with near-isometric distortion.","The distortion bound tends to $1+\\varepsilon$ as $\\lambda\\to\\infty$, so the discrete constraint becomes asymptotically costless in the large-scale limit.","The output lies in $\\frac{1}{\\lambda_0}\\mathbb{Z}^k$, which means the embedded vectors have bounded magnitude and fixed decimal precision, matching the constraints of a quantised computational model.","The dimension $k$ is independent of the ambient dimension $d$, so the reduction can be dramatic when $d\\gg\\log d$."],"supporting_citations":[{"why":"The classical Johnson–Lindenstrauss lemma supplies the base projection from $d$ to $k$ dimensions with distortion $1+\\varepsilon$ that the construction starts from.","marker":"[4]"},{"why":"Gives the form of the JL map $\\Phi(x)=\\frac{1}{\\sqrt{k}}Rx$ whose image lies in a lattice, which is needed for the argument.","marker":"[7]"},{"why":"Provides the same matrix representation of the JL transform, referenced alongside [7] for the structure of $\\Phi$.","marker":"[8]"},{"why":"Ziegler's theorem is the key tool guaranteeing a rotation that brings the dilated projected points close to the integer lattice.","marker":"[9]"},{"why":"Supplies the method for Lemma 2.1, showing that dilations of $t\\mathbb{Z}^k$ can be rotated near $\\mathbb{Z}^k$; used to construct the scale $\\lambda_1$.","marker":"[2]"},{"why":"Uniform distribution modulo one for the irrational sequence $(nt)$ underpins the existence of the scale $\\lambda_1$ in Lemma 2.1.","marker":"[6]"}],"fun_headline_variants":["Discrete JL: lattice points into ~log d dimensions","Lattice points flatten in O(log d) dimensions","Discrete subgroups: dimension reduction to log d","JL flattening for point sets on discrete grids"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof depends on comparing the upper bound measured on the dilated points $\\lambda x$ with the lower bound measured on the original points $x$, assuming these two measurements are on the same scale; if the factor $\\lambda$ separates them, the chained inequality cannot yield the claimed uniform distortion.","fun_headline_variants_meta":{"raw":{"variants":["Discrete JL: lattice points into ~log d dimensions","Lattice points flatten in O(log d) dimensions","Discrete subgroups: dimension reduction to log d","JL flattening for point sets on discrete grids"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000767,"raw_usage":{"total_tokens":3387,"prompt_tokens":918,"completion_tokens":2469,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":534,"completion_tokens_details":{"reasoning_tokens":2407}},"tokens_in":534,"tokens_out":2469,"duration_ms":17172,"temperature":1.0,"reasoning_tokens":2407,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:29:00.371775+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For $\\lambda_0=2$ and $\\varepsilon=1/4$, take two points in $D$ at the minimal lattice separation $\\lambda/\\lambda_0$ apart. Substituting the unscaled distance $\\|x_i-x_j\\|=1/\\lambda_0$ into the proof's inequality (2.3) reduces its last step to the requirement $\\varepsilon\\le \\varepsilon/\\lambda_0^2$, which is false for $\\lambda_0>1$; checking (1.1) for this pair therefore settles whether the claimed uniform distortion holds for all pairs.","supporting_citations":[{"cited_title":"Kuipers and H","cited_arxiv_id":null,"evidence_quote":"Gives the form of the JL map $\\Phi(x)=\\frac{1}{\\sqrt{k}}Rx$ whose image lies in a lattice, which is needed for the argument."},{"cited_title":"Matou s ek, On variants of the Johnson–Lindenstrauss lemma, Random Struct","cited_arxiv_id":null,"evidence_quote":"Provides the same matrix representation of the JL transform, referenced alongside [7] for the structure of $\\Phi$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Ziegler's theorem is the key tool guaranteeing a rotation that brings the dilated projected points close to the integer lattice."},{"cited_title":"Approximate embedding of large polygons into $Z^2$","cited_arxiv_id":"1208.1026","evidence_quote":"Supplies the method for Lemma 2.1, showing that dilations of $t\\mathbb{Z}^k$ can be rotated near $\\mathbb{Z}^k$; used to construct the scale $\\lambda_1$."},{"cited_title":"Kłopotek, Machine learning friendly set version of Johnson–Lindenstrauss lemma, Knowledge and Information Systems, 62(5) : 1961-2009, 2020","cited_arxiv_id":null,"evidence_quote":"Uniform distribution modulo one for the irrational sequence $(nt)$ underpins the existence of the scale $\\lambda_1$ in Lemma 2.1."}],"review_version":1}