{"paper_id":"2503.13193","title":"The deep multi-FBSDE method: a robust deep learning method for coupled FBSDEs","authors":["Kristoffer Andersson","Adam Andersson","Cornelis W. Oosterlee"],"paper_date":"2025-03-17","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2503.13193v3","page_url":"https://thequant.space/flowcharts/the-deep-multi-fbsde-method-a-robust-deep-learning-method-f/","scores":{"empirical_rigor":4,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes the deep multi-FBSDE method, a two‑phase deep‑learning algorithm for robust approximation of coupled forward‑backward stochastic differential equations.","quote":"We introduce the deep multi-FBSDE method for robust approximation of coupled forward-backward stochastic differential equations (FBSDEs), focusing on cases where the deep BSDE method of Han, Jentzen, and E (2018) fails to converge. Our algorithm proceeds in two phases: first, we approximate the initial condition for the FBSDE family, and second, we approximate the original FBSDE using the initial condition","summary":"We introduce the deep multi-FBSDE method for robust approximation of coupled forward-backward stochastic differential equations (FBSDEs), focusing on cases where the deep BSDE method of Han, Jentzen, and E (2018) fails to converge. To overcome the convergence issues, we consider a family of FBSDEs…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"A family of FBSDEs is considered that is equivalent to the original problem by satisfying the same associated PDE.","type":"model","stated":"explicit","quote":"To overcome the convergence issues, we consider a family of FBSDEs that are equivalent to the original problem in the sense that they satisfy the same associated partial differential equation (PDE).","support":"full"},{"assumption":"Regularity conditions on the coefficients of the stochastic control problem (b, σ, f, g) are assumed so that the HJB equation has a classical solution.","type":"model","stated":"explicit","quote":"(13) In this paper we tacitly assume such regularity.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Training data of Wiener increments","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"Training data con-sist of 220 independent realizations of the d-dimensional Wiener increment vectors ∆W0, . . . , ∆WN−1, each element being i.i.d., and normally distributed with mean zero and variance h.","support":"full"}],"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":"unclear","findings":[{"finding":"Numerical experiments show that the deep multi-FBSDE method converges even for problems where the standard deep BSDE method fails.","kind":"theoretical","quote":"Numerical experiments show that our method converges even when the standard deep BSDE method does not. Our numerical experiments demonstrate stable and accurate solutions for problems for which the classical deep FBSDE methods fail.","support":"full"},{"finding":"For simple low‑dimensional problems Phase I alone provides an accurate approximation, making Phase II unnecessary.","kind":"performance","quote":"(17) For this relatively straightforward problem, we have observed that only the first phase of the deep multi-FBSDE method is necessary, so Phase II is skipped.","support":"full"},{"finding":"The deep multi‑FBSDE method yields accurate approximations for all tested values of the parameter β, whereas the deep BSDE method fails for β > 0.25.","kind":"performance","quote":"The deep multi-FBSDE method provides accurate approximations for all β tested.","support":"partial"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":14,"citations":{"cited_by_in_corpus":1,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"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":2,"extracted_at":"2026-11-14T19:51:25Z"},"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/"}