An official website of the United States government
Conservation translocations, intentional movements of species to protect against extinction, have become widespread in recent decades and are projected to increase further as biodiversity loss continues worldwide. The literature abounds with analyses to inform translocations and assess whether they are successful, but the fundamental question of whether they should be initiated at all is rarely addressed formally. We used decision analysis to assess northern leopard frog reintroduction in northern Idaho, with success defined as a population that persists for at least 50 years. The Idaho Department of Fish and Game was the decision maker (i.e., the agency that will use this assessment to inform their decisions). Stakeholders from government, indigenous groups, academia, land management agencies, and conservation organizations also participated. We built an age-structured population model to predict how management alternatives would affect probability of success. In the model, we explicitly represented epistemic uncertainty around a success criterion (probability of persistence) characterized by aleatory uncertainty. For the leading alternative, the mean probability of persistence was 40%. The distribution of the modelling results was bimodal, with most parameter combinations resulting in either very low (<5%) or relatively high (>95%) probabilities of success. Along with other considerations, including cost, the Idaho Department of Fish and Game will use this assessment to inform a decision regarding reintroduction of northern leopard frogs. Conservation translocations may benefit greatly from more widespread use of decision analysis to counter the complexity and uncertainty inherent in these decisions. History: This paper has been accepted for the Decision Analysis Special Issue on Further Environmental Sustainability. Funding: This work was supported by the Wilder Institute/Calgary Zoo, the U.S. Fish and Wildlife Service [Grant F18AS00095], the NSF Idaho EPSCoR Program and the National Science Foundation [Grant OIA-1757324], and the Hunt Family Foundation. Supplemental Material: The online appendix is available at https://doi.org/10.1287/deca.2023.0472 .
more »
« less
Subjective Probability Correction for Uncertainty Representations
We propose a new approach to uncertainty communication: we keep the uncertainty representation fixed, but adjust the distribution displayed to compensate for biases in people’s subjective probability in decision-making. To do so, we adopt a linear-in-probit model of subjective probability and derive two corrections to a Normal distribution based on the model’s intercept and slope: one correcting all right-tailed probabilities, and the other preserving the mode and one focal probability. We then conduct two experiments on U.S. demographically-representative samples. We show participants hypothetical U.S. Senate election forecasts as text or a histogram and elicit their subjective probabilities using a betting task. The first experiment estimates the linear-in-probit intercepts and slopes, and confirms the biases in participants’ subjective probabilities. The second, preregistered follow-up shows participants the bias-corrected forecast distributions. We find the corrections substantially improve participants’ decision quality by reducing the integrated absolute error of their subjective probabilities compared to the true probabilities. These corrections can be generalized to any univariate probability or confidence distribution, giving them broad applicability. Our preprint, code, data, and preregistration are available at https://doi.org/10.17605/osf.io/kcwxm
more »
« less
- PAR ID:
- 10504923
- Publisher / Repository:
- ACM
- Date Published:
- Journal Name:
- CHI '23: Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems
- ISBN:
- 9781450394215
- Page Range / eLocation ID:
- 1 to 17
- Format(s):
- Medium: X
- Location:
- Hamburg Germany
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Traditional deep neural networks (NNs) have significantly contributed to the state-of-the-art performance in the task of classification under various application domains. However, NNs have not considered inherent uncertainty in data associated with the class probabilities where misclassification under uncertainty may easily introduce high risk in decision making in real-world contexts (e.g., misclassification of objects in roads leads to serious accidents). Unlike Bayesian NN that indirectly infer uncertainty through weight uncertainties, evidential NNs (ENNs) have been recently proposed to explicitly model the uncertainty of class probabilities and use them for classification tasks. An ENN offers the formulation of the predictions of NNs as subjective opinions and learns the function by collecting an amount of evidence that can form the subjective opinions by a deterministic NN from data. However, the ENN is trained as a black box without explicitly considering inherent uncertainty in data with their different root causes, such as vacuity (i.e., uncertainty due to a lack of evidence) or dissonance (i.e., uncertainty due to conflicting evidence). By considering the multidimensional uncertainty, we proposed a novel uncertainty-aware evidential NN called WGAN-ENN (WENN) for solving an out-of-distribution (OOD) detection problem. We took a hybrid approach that combines Wasserstein Generative Adversarial Network (WGAN) with ENNs to jointly train a model with prior knowledge of a certain class, which has high vacuity for OOD samples. Via extensive empirical experiments based on both synthetic and real-world datasets, we demonstrated that the estimation of uncertainty by WENN can significantly help distinguish OOD samples from boundary samples. WENN outperformed in OOD detection when compared with other competitive counterpartsmore » « less
-
This study employs a spatial autoregressive probit-log linear (SAP-Log) hurdle model to investigate the influencing factors on the probability of death and case fatality rate (CFR) of coronavirus disease 2019 (COVID-19) at the city level in China. The results demonstrate that the probability of death from COVID-19 and the CFR level are 2 different processes with different determinants. The number of confirmed cases and the number of doctors are closely associated with the death probability and CFR, and there exist differences in the CFR and its determinants between cities within Hubei Province and outside Hubei Province. The spatial probit model also presents positive spatial autocorrelation in death probabilities. It is worth noting that the medical resource sharing among cities and enjoyment of free medical treatment services of citizens makes China different from other countries. This study contributes to the growing literature on determinants of CFR with COVID-19 and has significant practical implications.more » « less
-
Individual probabilities refer to the probabilities of outcomes that are realized only once: the probability that it will rain tomorrow, the probability that Alice will die within the next 12 months, the probability that Bob will be arrested for a violent crime in the next 18 months, etc. Individual probabilities are fundamentally unknowable. Nevertheless, we show that two parties who agree on the data—or on how to sample from a data distribution—cannot agree to disagree on how to model individual probabilities. This is because any two models of individual probabilities that substantially disagree can together be used to empirically falsify and improve at least one of the two models. This can be efficiently iterated in a process of “reconciliation” that results in models that both parties agree are superior to the models they started with, and which themselves (almost) agree on the forecasts of individual probabilities (almost) everywhere. We conclude that although individual probabilities are unknowable, they are contestable via a computationally and data efficient process that must lead to agreement. Thus we cannot find ourselves in a situation in which we have two equally accurate and unimprovable models that disagree substantially in their predictions—providing an answer to what is sometimes called the predictive or model multiplicity problem.more » « less
-
This study proposes a multi-period facility location formulation to maximize coverage while meeting a coverage reliability constraint. The coverage reliability constraint is a chance constraint limiting the probability of failure to maintain the desired service standard, commonly followed by emergency medical services and fire departments. Further, uncertainties in the failure probabilities are incorporated by utilizing robust optimization using polyhedral uncertainty sets, which results in a compact mixed-integer linear program. A case study in the Portland, OR metropolitan area is analyzed for employing unmanned aerial vehicles (UAVs) or drones to deliver defibrillators in the region to combat out-of-hospital cardiac arrests. In the context of this study, multiple periods represent periods with different wind speed and direction distributions. The results show that extending to a multi-period formulation, rather than using average information in a single period, is particularly beneficial when either response time is short or uncertainty in failure probabilities is not accounted for. Accounting for uncertainty in decision-making improves coverage significantly while also reducing variability in simulated coverage, especially when response times are longer. Going from a single-period deterministic formulation to a multi-period robust formulation boosts the simulated coverage values by 57%, on average. The effect of considering a distance-based equity metric in decision-making is also explored.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3544548.3580998
