Certified Probability Landscapes for Stochastic Reaction Networks

Definitions and mathematical statement

A probability distribution on a measurable state space $(S,\Sigma)$ is a countably additive measure $\mu:\Sigma\to[0,1]$ with $\mu(S)=1$. For an event $A\in\Sigma$, its probability is $\mu(A)$. These definitions provide notation only: the source does not specify the state space, reaction network, stochastic dynamics, initial conditions, or target distribution. In particular, it does not say whether the intended probability landscape is a transient distribution, a stationary distribution, a density, or a transformation of a distribution.

The supplied research problem asks, for stochastic chemical reaction networks, how to certify the accuracy of a computed probability landscape, determine the best attainable accuracy for laptop-scale and large-scale computation, and control errors including missed probability peaks and errors in rare-event probabilities.

A precise mathematical assertion cannot be extracted without further specifications. The source gives no class of admissible networks, computational approximation, error functional, tolerance, confidence requirement, or resource budget. It also supplies no definition of a probability peak, criterion for detecting all relevant peaks, or threshold defining a rare event. Absolute and relative errors in event probabilities are distinct possible targets, and the source does not select either. Consequently, the requested certification, optimal accuracy, and error control remain research objectives rather than a fully quantified theorem or conjecture. No additional hypotheses or quantifiers are imposed here.

Short English statement

The problem asks how to certify numerical probability landscapes for stochastic chemical reaction networks, compare attainable accuracy at laptop and larger computational scales, and control missed peaks and rare-event errors. The target landscape, error criterion, network class, and computational budgets are unspecified, so the supplied record does not determine a unique precise mathematical claim.

Sources

• Local UnsolvedMath registry, permanent ID $30003738$, title: Certified Probability Landscapes for Stochastic Reaction Networks; source problem code: catalogue entry. The supplied record belongs to the Oberwolfach Reports Open Problems collection and records proposed year 2018. No individual proposer is identified.
• Supplied report citation: The Mathematics of Mechanobiology and Cell Signaling (2019), Oberwolfach Reports, pp. 433--506, DOI 10.4171/owr/2018/8. Supplied source link: https://doi.org/10.4171/owr/2018/8. The background describes Oberwolfach Reports as documenting mathematical workshops and their open-problem discussions.
• Supplied literature source, labeled "original OWR report": https://doi.org/10.4171/OWR/2018/8.
• Literature assessment dated exactly 2026-08-21: the supplied curation labels the problem open and states that no single mathematical claim is identifiable because the probability landscape, estimator, discretization, and error criterion are unspecified. This is supplied dated curation, not an independently verified literature assessment. The background requests independent verification of time-sensitive status and cited claims.
• The registry marks the statement as "exact" and describes it as clear, self-contained, publication-ready, and retained byte-for-byte. That supplied verification label does not resolve the missing mathematical specifications identified above. The report and its attribution have not been independently checked for this draft.