{"paper_id":"2305.16274","title":"Non-adversarial training of Neural SDEs with signature kernel scores","authors":["Zacharia Issa","Blanka Horvath","Maud Lemercier","Cristopher Salvi"],"paper_date":"2023-05-25","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2305.16274v1","page_url":"https://thequant.space/flowcharts/non-adversarial-training-of-neural-sdes-with-signature-kerne/","scores":{"empirical_rigor":3,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a non‑adversarial training method for Neural SDEs using signature kernel scores as the objective.","quote":"we introduce a novel class of scoring rules on pathspace based on signature kernels and use them as objective for training Neural SDEs non-adversarially.","summary":"Neural SDEs are continuous-time generative models for sequential data. State-of-the-art performance for irregular time series generation has been previously obtained by training these models adversarially as GANs. However, as typical for GAN architectures, training is notoriously unstable, often…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Technical assumption needed to ensure characteristicness of the signature kernel.","type":"model","stated":"explicit","quote":"This is a technical assumption needed to ensure characteristicness of the signature kernel. See Proposition (3.1)."},{"assumption":"Paths x, y are almost‑everywhere differentiable so that the signature kernel satisfies a linear hyperbolic PDE.","type":"model","stated":"explicit","quote":"which reduces to a linear hyperbolic PDE in the when the paths x, y are almost-everywhere dif‑ferentiable."},{"assumption":"The space X is equipped with a topology that makes the signature continuous.","type":"model","stated":"explicit","quote":"From now on we endow X with a topology with the respect to which the signature is continuous; see [CT22] for various choices of such topologies."},{"assumption":"A target random variable Xtrue with law PXtrue is available for training.","type":"data","stated":"explicit","quote":"We are given a target X-valued random variable Xtrue with law PXtrue."}]},"datasets":{"status":"available","items":[{"name":"Geometric Brownian motion synthetic dataset","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"We chose µ = 0, σ = 0.2 and generated time-augmented paths of length 64 over the grid ∆= {0, 1, . . . , 63} with dt = 0.01."},{"name":"Foreign exchange (FX) currency pairs dataset","provider":null,"kind":"market-prices","access":"unknown","period":null,"universe":null,"quote":"the third is a multidimensional example with foreign exchange (forex, or FX) currency pairs, which was chosen not only because of the relevance and capitalisation of FX markets but also due to its well-known intricate complexity."},{"name":"NASDAQ limit order book data","provider":"NASDAQ public exchange [NMK+18]","kind":"order-book","access":"public","period":null,"universe":null,"quote":"We fit our model on real LOB data from the NASDAQ public exchange [NMK+18] which consists of about 4M timestamped events with L = 10 price levels."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"l2-book"},"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":"forecast-accuracy","findings":[{"finding":"The generator trained with the full signature kernel score (ϕsig) achieves the lowest KS scores and Type I error at the 5 % significance level on the geometric Brownian motion benchmark.","kind":"performance","quote":"Here the generator trained with ϕsig performs the best, achieving a Type I error at the assumed confidence level."},{"finding":"The proposed non‑adversarial training significantly outperforms alternative training methods (including SDE‑GAN) across multiple tasks such as rough volatility simulation, FX forecasting, and limit order book generation.","kind":"performance","quote":"significantly outperforms alternative ways of training Neural SDEs on a variety of tasks including the simulation of rough volatility models, the conditional probabilistic forecasts of real‑world forex pairs"},{"finding":"Average KS scores for each spatiotemporal marginal of the NASDAQ limit order book are reported, showing competitive performance of the Neural SPDE model trained with signature kernel scores.","kind":"performance","quote":"Figure 3: KS test average scores for each spatiotemporal marginal, 100 runs, NASDAQ data."}],"reported_performance":null,"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[],"contradicts":[],"qualifies":[],"compared_with":15,"citations":{"cited_by_in_corpus":6,"by_independent_authors":5,"how_cited":{"used_as_baseline":0,"uses_method_or_data":1,"extends":0,"discusses_findings":0,"mentions":3,"unclassified":2},"independent_papers":[{"paper":{"paper_id":"2406.10214","title":"Universal randomised signatures for generative time series modelling","paper_date":"2024-06-14","authors":["Francesca Biagini","Lukas Gonon","Niklas Walter"],"empirical_rigor":7.5,"page_url":"https://thequant.space/flowcharts/universal-randomised-signatures-for-generative-time-series-modelling/"},"how":"uses_method_or_data","quote":null}],"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":"2026-11-14T18:26:43Z"},"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/"}