{"id":"996f5616-995f-4569-8680-1a695f56192f","arxiv_id":"1908.07676","paper_version":4,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"For interval maps with an odd-periodic point, transitivity is equivalent to transitivity of the induced map on probability measures, and several counterexamples separate other dynamical properties.","lead":"Mathematicians studied what happens to the dynamics of a continuous map when it is lifted to the space of probability measures over its state space. This paper pins down when transitivity, mixing, sensitivity, and chaotic properties are preserved under that lift, especially for maps on an interval and for time-varying systems.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the interval transitivity criterion is internally sound, and the only defect is an imprecise proof sentence in Theorem 2.3(i) that omits the odd-period qualifier.","rationale":"The reader's weakest assumption identifies reliance on Ruette's external classification and the missing qualifier in the proof of Theorem 2.3(i). I agree the sentence is imprecise and that the classification is external, but I do not regard this as a load-bearing correctness risk: the theorem statement includes exactly the qualifier that the proof sentence omits, and Ruette's Theorem 2.20 is a standard published result whose hypothesis matches the paper's use. The rest of the proof, including the independent proof of Theorem 2.11 and the explicit counterexample in Theorem 2.3(ii), is internally consistent. Since the concern is a fixable presentation issue rather than a mathematical gap, I would not move the verdict to conditional on mathematical grounds; leaving the reader's conditional verdict unchanged reflects the minor presentational caveat while recording that no substantive objection surfaced.","tokens_in":32351,"tokens_out":28901,"duration_ms":251189,"concrete_test":"Verify the exact statement of Ruette [23, Theorem 2.20]: confirm that a transitive interval map with a periodic point of odd period different from 1 is mixing, and then insert the qualifier 'under the assumption of (i)' into the proof of Theorem 2.3(i). If Ruette's theorem instead carries an extra hypothesis, that hypothesis must be imported explicitly; otherwise the proof as intended is complete.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is Theorem 2.3. The proof has two load-bearing steps: Proposition 2.2 gives M-transitivity implies I-transitivity, and the converse uses Ruette's Theorem 2.20 to promote interval transitivity under the odd-period hypothesis to mixing, then uses the paper's own Theorem 2.11 to lift mixing to M(I). I checked the proof chain: Theorem 2.11 is proved independently via Lemma 2.8, and the counterexample in Theorem 2.3(ii) has an explicit ball-separation argument that is consistent with the statement. The only soft spot is a proof sentence that says 'transitivity of (I,f) is equivalent to mixing of (I,f) by Theorem 2.20' without repeating the odd-period qualifier. Read literally this equivalence is false for the map in part (ii), but the theorem's hypothesis supplies the qualifier and Ruette's result is a standard interval-map classification. This is a syntactical imprecision, not a gap in the mathematics. I found no other load-bearing concern; the entropy, chain, shadowing, and sensitivity arguments are internally consistent and appropriately cited.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the relations between the dynamics of a continuous map (or non-autonomous system) on a compact metric space X and the dynamics of the induced pushforward map on the space M(X) of Borel probability measures with the Prohorov metric. The principal result is Theorem 2.3: for an interval map (I,f) that has a periodic point of odd period different from 1, (I,f) is transitive if and only if (M(I),\\hat f) is transitive, and a counterexample shows that the odd-period condition is sharp. The paper also proves equivalence of mixing, mild mixing, exactness, and several forms of sensitivity between the original and induced systems, studies topological entropy, chain mixing and chain transitivity, shadowing and specification, weak mixing of higher orders, Li-Yorke and distributional chaos, and equi-conjugacy. A number of results are extended to non-autonomous systems, with examples separating autonomous and non-autonomous behavior.","tokens_in":32408,"tokens_out":23394,"duration_ms":212316,"significance":"If the results hold, the paper gives a clean, sharp criterion for the transitivity equivalence on intervals and substantially extends the existing theory of induced measure systems to non-autonomous dynamics. The main strength is the combination of general transfer results with carefully constructed counterexamples, such as the ball-separation argument in Theorem 2.3(ii), the two-point examples for chain mixing and shadowing, and the entropy example in Theorem 2.6. The reliance on standard external classifications (Ruette's interval theorem, Balibrea-Oprocha's weak mixing result) is transparent and does not appear to be circular. The paper is a solid contribution to the induced-measure-space dynamics literature, provided the technical issues below are corrected.","major_comments":[{"comment":"The proof states: \"transitivity of (I,f) is equivalent to mixing of (I,f) by Theorem 2.20 in [23].\" As written, this equivalence is false for the map constructed in part (ii) of the same theorem, which is transitive but not mixing. The intended statement is that, under the hypothesis that (I,f) has a periodic point of odd period different from 1, transitivity is equivalent to mixing. The proof should explicitly repeat the odd-period qualifier to avoid a logically false intermediate claim.","section":"Section 2.1, proof of Theorem 2.3(i)"},{"comment":"The measure \\hat\\nu defined in (4.11) is written as (1/|A|)\\sum_{j\\in A}\\delta_{y_j} + (1/(n_0-|A|))\\sum_{j\\notin A}\\delta_{x_j}. This does not define a probability measure: the coefficients sum to 2, and the expression is undefined when |A|=n_0. The same problem occurs in the definition of \\hat\\nu_i in the proof of Theorem 4.4. The intended weights are almost certainly 1/n_0 (and 1/n_i respectively), and with that correction the subsequent inequalities are valid. As printed, the proofs are invalid and must be corrected.","section":"Section 4, Theorem 4.3, Eq. (4.11), and Theorem 4.4"},{"comment":"The proof invokes \"Lemma 3.7\" to assert that \\hat f_0^k is surjective, but no Lemma 3.7 exists in the manuscript. The intended reference appears to be Lemma 3.2, applied iteratively to each surjective map f_i, which gives surjectivity of each \\hat f_i and hence of the composition. This citation error should be fixed.","section":"Section 3.1, proof of Theorem 3.3"}],"minor_comments":[{"comment":"The equality \\hat f_0^{n_0}(\\overline{B_{P_d}(\\delta_x,r)}) = \\hat f_0^{n_0}(D) is not literally correct; by continuity and compactness the left side equals the closure of \\hat f_0^{n_0}(D). The intended conclusion about the distance from \\delta_y to the image of the closed ball still follows, but the equality should be replaced by the appropriate closure statement.","section":"Section 2.3, proof of Theorem 2.12"},{"comment":"The forward direction of the uniform convergence proof only writes the inequality \\hat f_n(\\mu)(A) \\le \\hat f(\\mu)(A^\\varepsilon)+\\varepsilon. The symmetric inequality needed for the Prohorov metric should also be stated or explicitly derived from the same uniform convergence assumption.","section":"Section 2.1, Lemma 2.5"},{"comment":"The proof applies the shadowing property to a finite \\delta_0-chain, while the definition of shadowing concerns infinite pseudo-orbits. The argument should note that a finite chain can be extended to an infinite \\delta_0-pseudo orbit, for instance by concatenating chains supplied by Theorem 3.3, before shadowing is applied.","section":"Section 3.2, proof of Theorem 3.9"},{"comment":"There are several typographical slips, such as \"exsits\" in the definition of sensitivity in Section 4 and \"Denfine\" in the proof of Theorem 4.4. These should be corrected in the final version.","section":"General"}],"recommendation":"minor_revision","confidential_remarks":"The central claim (Theorem 2.3) is sound under external standard hypotheses, and the remaining issues are local typographical or presentational errors. The measure weight error in Theorems 4.3 and 4.4 is easily fixed and does not affect the underlying idea. The paper is within the scope of the journal and, with the corrections, would be a worthwhile contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The two things to know: the sharp interval criterion in Theorem 2.3 is real, and the paper closes several open questions from Bauer-Sigmund and the later literature. The non-autonomous weak-mixing counterexample (Theorem 2.7) is the most interesting new construction; it shows the autonomous equivalence genuinely breaks down outside the single-map setting. The chain and shadowing results (Theorems 3.3, 3.9, 3.10) also look solid and are a genuine extension of Bernardes-Vermersch.\n\nThe paper does a lot right. The proofs are detailed and the main constructions are explicit enough to check. The comparison with prior work is honest, and the paper identifies exactly which questions it closes. Theorem 2.3(i) is the centerpiece: under the odd-period condition, interval transitivity is equivalent to transitivity of the induced measure map, and the example in part (ii) shows the condition is necessary. I checked the proof chain and it holds; the key lifting step (Theorem 2.11) is proved independently, and the counterexample has a real ball-separation argument.\n\nThe soft spots are minor but worth naming. The proof of Theorem 2.3(i) contains a sentence that says transitivity of (I,f) is equivalent to mixing of (I,f) without repeating the odd-period qualifier. Read literally that is false for the map in part (ii). The theorem hypothesis supplies the missing qualifier and Ruette's classification is standard, so this is a syntactical slippage, not a gap. The paper also imports several external results as black boxes: Ruette's interval classification, Kolyada-Snoha on non-autonomous entropy, and Balibrea-Oprocha for weak mixing of the constructed non-autonomous system. Those citations look appropriate, but the paper would be stronger if it stated the exact invoked versions. There are also assorted typos (\"exsits\", \"reps.\", \"Denfine\") that an editor should let the authors clean up.\n\nOverall the mathematics is internally consistent and the claims are as advertised. For a specialist in topological dynamics, this is a useful paper that settles open questions and gives honest counterexamples. The defects do not touch the load-bearing arguments. A serious referee should engage with it, and with light revision the paper is publishable.\n\nRecommendation: accept for refereeing; the authors should fix the qualifier in the proof sentence and tighten the external-theorem statements.","headline":"The interval transitivity criterion is a genuine sharp result and the non-autonomous counterexamples are worth having; the paper deserves serious refereeing despite a few proof-writing blemishes.","tokens_in":33085,"tokens_out":1063,"would_cite":true,"duration_ms":574147,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37A50","54H20","37B55","60B05"],"pacs":[],"model":"deepseek-v4-flash","headline":"For interval maps with a periodic point of odd period, transitivity of the map and of its induced probability-measure map are equivalent.","keywords":["induced map on probability measures","topological transitivity","mixing","chain mixing","shadowing","Li-Yorke chaos","non-autonomous dynamical systems","interval maps"],"falsifier":"Exhibit a continuous interval map with a periodic point of odd period greater than 1 and a dense orbit whose induced map on $\\mathcal{M}(I)$ is not topologically transitive; Theorem 2.3 says no such map exists.","tokens_in":32004,"feed_emoji":"🔄","tokens_out":9030,"duration_ms":80792,"temperature":0.7,"pith_summary":"This paper studies how the dynamics of a continuous map on a compact space are reflected by the map it induces on the space of Borel probability measures. Its headline result is a sharp interval criterion: if $f$ has a periodic point of odd period different from 1, then $(I,f)$ is transitive if and only if $(\\mathcal{M}(I),\\hat f)$ is transitive. It also proves that sensitivity, total transitivity, mixing, mild mixing, and exactness pass between the base system and the measure system under the stated conditions, and that transitive interval systems force infinite topological entropy on the measure space. A separate thread shows that non-autonomous systems behave differently, with weak mixing of order 2 and entropy failing to lift in general.","feed_headline":"Odd-period points decide when induced maps stay transitive","feed_subtitle":"Transitivity and sensitivity of an interval map and of its probability-measure system coincide under a sharp condition.","key_machinery":"The central machinery is the pushforward operator $\\hat f$ on $\\mathcal{M}(X)$, together with the Prohorov metric making $\\mathcal{M}(X)$ compact. Most proofs run through two approximation lemmas: the set of finitely supported measures $\\frac{1}{n}\\sum_{i=1}^{n}\\delta_{x_i}$ is dense in $\\mathcal{M}(X)$, and any finite family of open subsets of $\\mathcal{M}(X)$ can be entered by such measures whose atoms lie in prescribed open subsets of $X$. These lemmas allow mixing, weak mixing, exactness, chain properties, and sensitivity to be transferred back and forth between the base dynamics and the measure dynamics.","core_discovery":"The paper's central claim is a sharp transitivity criterion for the induced system. For a continuous self-map $f$ of a compact interval $I$, write $\\hat f$ for the map on the space $\\mathcal{M}(I)$ of Borel probability measures defined by $\\hat f(\\mu)(A)=\\mu(f^{-1}(A))$. Theorem 2.3 states that if $f$ has a periodic point of odd period different from 1, then $(I,f)$ is transitive if and only if $(\\mathcal{M}(I),\\hat f)$ is transitive; the theorem also exhibits a piecewise-linear transitive interval map with no such periodic point whose induced system is not transitive, so the condition cannot be relaxed. Around this criterion the paper establishes equivalences for total transitivity, mixing, mild mixing, exactness, and several sensitivity notions on intervals, along with a one-way inheritance of Li-Yorke and distributional chaos in the general compact metric setting.","pith_inferences":["The odd-period criterion suggests a usable test in wider classes of spaces: failure of transitivity to lift to the measure space may be driven by absence of nontrivial odd periodic structure rather than by compactness or dimension alone.","The zero-entropy non-autonomous example indicates that time-variation can erase the entropy amplification that autonomous transitive intervals display; a natural extension is to decide exactly which non-autonomous sequences restore infinite entropy on the measure space.","The chain-mixing result for surjective sequences, paired with the two counterexamples, implies that non-surjectivity is the main obstruction to the converse; a sharper converse would probably characterize chain-mixing induced measure systems among non-surjective maps."],"forward_implications":["On intervals, any transitive map with a periodic point of odd period greater than 1 has a transitive induced measure system, so the measure space is not merely a passive recorder of the base dynamics.","Sensitive interval maps induce sensitive measure systems; the paper proves the stronger conclusion that cofinite sensitivity transfers, which also covers syndetic and ergodic sensitivity.","Every transitive interval map forces infinite topological entropy on $\\mathcal{M}(I)$, so finite entropy on the measure side is impossible for transitive autonomous interval systems.","Chain transitivity of a compact system makes its induced measure system chain mixing, chain weakly mixing of all orders, chain exact, and chain transitive; surjectivity of every map in a non-autonomous sequence suffices for the same conclusion.","If a surjective non-autonomous system is not mixing, its induced measure system does not have shadowing, so shadowing is not automatically inherited by the measure space."],"supporting_citations":[{"why":"introduces the induced measure system and proves that transitivity of the induced map implies transitivity of the base map.","marker":"[2]"},{"why":"supplies the classification that transitive interval maps with a periodic point of odd period are mixing, the backbone of Theorem 2.3.","marker":"[23]"},{"why":"provides the non-autonomous weak-mixing-of-order-2 example used to show that weak mixing need not lift to the measure space.","marker":"[1]"},{"why":"gives the one-sided inequality for the Prohorov metric used throughout the measure-space estimates.","marker":"[27]"},{"why":"supplies prior results on induced measure systems, including chain mixing for homeomorphisms and a no-Li-Yorke-pair example that the paper extends or contrasts.","marker":"[5]"},{"why":"yields the positive lower bound for topological entropy of transitive interval maps used to prove infinite entropy of the induced system.","marker":"[7]"},{"why":"defines topological sequence entropy for non-autonomous systems and supplies the bound used to show the zero-entropy example.","marker":"[12]"},{"why":"supplies the fact that chain transitivity implies surjectivity, used in the chain-mixing corollary.","marker":"[14]"}],"fun_headline_variants":["Odd-period points pin down induced map transitivity","Transitivity and sensitivity match for measure maps","Measure maps inherit chaos but not shadowing","Sharp criterion: odd-period points ensure transitivity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof of the main interval criterion relies on an imported classification theorem stating that, under the odd-period condition, transitivity of an interval map implies mixing; if that classification is wrong or inapplicable where it is quoted, the transitivity equivalence would need another argument.","fun_headline_variants_meta":{"raw":{"variants":["Odd-period points pin down induced map transitivity","Transitivity and sensitivity match for measure maps","Measure maps inherit chaos but not shadowing","Sharp criterion: odd-period points ensure transitivity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000257,"raw_usage":{"total_tokens":1718,"prompt_tokens":1227,"completion_tokens":491,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":843,"completion_tokens_details":{"reasoning_tokens":434}},"tokens_in":843,"tokens_out":491,"duration_ms":5066,"temperature":1.0,"reasoning_tokens":434,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:02:37.806074+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a continuous interval map with a periodic point of odd period greater than 1 and a dense orbit whose induced map on $\\mathcal{M}(I)$ is not topologically transitive; Theorem 2.3 says no such map exists.","supporting_citations":[{"cited_title":"Bauer, K","cited_arxiv_id":null,"evidence_quote":"introduces the induced measure system and proves that transitivity of the induced map implies transitivity of the base map."},{"cited_title":"Ruette, Chaos on the interval, University Lecture Se ries, 67","cited_arxiv_id":null,"evidence_quote":"supplies the classification that transitive interval maps with a periodic point of odd period are mixing, the backbone of Theorem 2.3."},{"cited_title":"Balibrea, P","cited_arxiv_id":null,"evidence_quote":"provides the non-autonomous weak-mixing-of-order-2 example used to show that weak mixing need not lift to the measure space."},{"cited_title":"Strassen, The existence of probability measures wit h given marginals, Ann","cited_arxiv_id":null,"evidence_quote":"gives the one-sided inequality for the Prohorov metric used throughout the measure-space estimates."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"supplies prior results on induced measure systems, including chain mixing for homeomorphisms and a no-Li-Yorke-pair example that the paper extends or contrasts."},{"cited_title":"Block, E","cited_arxiv_id":null,"evidence_quote":"yields the positive lower bound for topological entropy of transitive interval maps used to prove infinite entropy of the induced system."},{"cited_title":"Kolyada, L","cited_arxiv_id":null,"evidence_quote":"defines topological sequence entropy for non-autonomous systems and supplies the bound used to show the zero-entropy example."},{"cited_title":"Kurka, Topological and symbolic dynamics, Cours sp´ ecialis´ es [specialized courses], Soci´ et´ e math´ ematique de France, Paris, 11, 2013","cited_arxiv_id":null,"evidence_quote":"supplies the fact that chain transitivity implies surjectivity, used in the chain-mixing corollary."}],"review_version":1}