{"paper_id":"2510.19494","title":"Quantum Machine Learning methods for Fourier-based distribution estimation with application in option pricing","authors":["Fernando Alonso","Álvaro Leitao","Carlos Vázquez"],"paper_date":"2025-10-22","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2510.19494v1","page_url":"https://thequant.space/flowcharts/quantum-machine-learning-methods-for-fourier-based-distribut/","scores":{"empirical_rigor":4,"math_complexity":8.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes two hybrid classical‑quantum methods that reconstruct Fourier series of statistical distributions using outputs of quantum machine learning models based on parametrized quantum circuits for option pricing.","quote":"In this work, we introduce two hybrid classical‑quantum methods to address the option pricing problem. These approaches rely on reconstructing Fourier series representations of statistical distributions from the outputs of Quantum Machine Learning (QML) models based on Parametrized Quantum Circuits (PQCs).","summary":"The ongoing progress in quantum technologies has fueled a sustained exploration of their potential applications across various domains. One particularly promising field is quantitative finance, where a central challenge is the pricing of financial derivatives-traditionally addressed through Monte…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"There exists a finite interval [a,b] such that the PDF tails vanish and the integral can be approximated on this interval.","type":"market","stated":"explicit","quote":"In finance, it is common to work in practice with PDFs whose tails tend to vanish, so it can be assumed that there exists an interval [a, b] ⊂R such that the integral in (4) can be approximated without significant loss of accuracy"},{"assumption":"The probability density can be approximated by a K‑truncated trigonometric Fourier series.","type":"model","stated":"explicit","quote":"Since f(y|x) is usually not known explicitly, we can approximate the density by its K‑truncated trigonometric Fourier series expansion in y"},{"assumption":"Trainable blocks in the PQC are arbitrary unitary operations.","type":"model","stated":"explicit","quote":"For simplicity, in the subsequent developments, we will assume that the trainable blocks are arbitrary unitary operations, i.e., W(θ) = W"},{"assumption":"The risk‑neutral measure is used for pricing under the expectation.","type":"market","stated":"explicit","quote":"In the literature, it is common to refer to it as the risk‑neutral measure."},{"assumption":"Fourier series of period 2(b−a) eliminates the Gibbs phenomenon for the CDF approximation.","type":"statistical","stated":"explicit","quote":"the Gibbs phenomenon is completely eliminated, because we are only considering the part of the series corresponding to the CDF"},{"assumption":"Increasing the expressive capacity of the PQC (circuit dimensions) improves convergence toward the true derivative value.","type":"model","stated":"implied","quote":"Both methods exhibited stable convergence towards the true derivative value as the expressive capacity of the PQCs increased."}]},"datasets":{"status":"available","items":[{"name":"European vanilla put option pricing dataset","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"For our numerical experiments, we consider the pricing of a European vanilla put option at t0 = 0, so that the payoff takes the form (see [1], for example): h(T, ST ) = max{K −ST , 0}."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"pricing-calibration","findings":[{"finding":"Both hybrid methods converge stably toward the true option price as PQC size increases, with circuit dimensions 6×6 to 8×8 for Method I and 4×4 to 6×6 for Method II yielding reliable estimates.","kind":"performance","quote":"Both methods exhibited stable convergence towards the true derivative value as the expressive capacity of the PQCs increased. Circuit dimensions on the order of 6 × 6 to 8 × 8 for Method I and 4 × 4 to 6 × 6 for Method II yielded consistent and reliable estimates"},{"finding":"Method III shows substantial improvement when the number of Fourier coefficients increases, especially with low numbers of shots, leading to refined price estimates.","kind":"performance","quote":"In the case of Method III, increasing the number of coefficients leads to a substantial improvement in the results, particularly when the number of shots is low."},{"finding":"Intermediate PQC schemes (7×7 for Method I and 5×5 for Method II) achieve the best performance, illustrating a trade‑off between model complexity and problem characteristics.","kind":"performance","quote":"the best performance is obtained with an intermediate scheme (7 × 7 for Method I and 5 × 5 for Method II), illustrating how in this type of problem there is often a trade‑off between model complexity and the specific characteristics of the problem"},{"finding":"Increasing dataset size leads to clear convergence toward the exact option price for Methods I and II, though occasional outliers appear.","kind":"performance","quote":"From Figure 5, a clear convergence towards the exact value in both Method I and Method II can be observed as the sample size increases. However, some outliers appear"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":35,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:23:54Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}