Geospatial uncertainty modeling using Stacked Gaussian Processes

Abdelfatah, Kareem; Bao, Junshu; Terejanu, Gabriel

doi:10.1016/j.envsoft.2018.08.022

Cited by 10 publications

(13 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many researchers used LGCPs for occurrence modeling in a wide range of contexts; see, for example, Serra et al (2012), Møller and Díaz-Avalos (2010), and Gabriel et al (2017). Furthermore, Abdelfatah et al (2016) and Trucchia et al (2018) used GPs for modeling burnt area.…”

Section: Introductionmentioning

confidence: 99%

A combined statistical and machine learning approach for spatial prediction of extreme wildfire frequencies and sizes

Daniela¹,

Gong²,

Yadav³

et al. 2021

Preprint

View full text Add to dashboard Cite

Motivated by the Extreme Value Analysis 2021 (EVA 2021) data challenge we propose a method based on statistics and machine learning for the spatial prediction of extreme wildfire frequencies and sizes. This method is tailored to handle large datasets, including missing observations. Our approach relies on a four-stage high-dimensional bivariate sparse spatial model for zero-inflated data, which is developed using stochastic partial differential equations (SPDE). In Stage 1, the observations are categorized in zero/nonzero categories and are modeled using a two-layered hierarchical Bayesian sparse spatial model to estimate the probabilities of these two categories. In Stage 2, before modeling the positive observations using a spatiallyvarying coefficients, smoothed parameter surfaces are obtained from empirical estimates using fixed rank kriging. This approximate Bayesian method inference was employed to avoid the high computational burden of large spatial data modeling using spatially-varying coefficients. In Stage 3, the standardized log-transformed positive observations from the second stage are further modeled using a sparse bivariate spatial Gaussian process. The Gaussian distribution assumption for wildfire counts developed in the third stage, is computationally effective but erroneous. Thus in Stage 4, the predicted values are rectified using Random Forests. Posterior inference is drawn for Stages 1 and 3 using Markov chain Monte Carlo (MCMC) sampling. A cross-validation scheme is then created for the artificially generated gaps, and the EVA 2021 prediction scores of the proposed model are compared to those obtained using certain natural competitors.

show abstract

Section: Introductionmentioning

confidence: 99%

A combined statistical and machine learning approach for spatial prediction of extreme wildfire frequencies and sizes

Daniela¹,

Gong²,

Yadav³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In this example, the optimal policy is to select action 𝑎 1 from 𝑠 1 , and the action selected from 𝑠 2 is irrelevant. When performing unconditional planning from state 𝑠 0 , we would find the optimal action sequence to be [𝑎 1 , 𝑎 1 ]. The set of observations with probability at least 0.5 that results in the highest conditional expectation is 𝒰 =…”

Section: Rescaling Of Prior Specializations In Online Planningmentioning

confidence: 99%

“…Now assume that after executing 𝑎 1 from 𝑠 0 , the agent observes o 1 1 and enters 𝑠 1 . If we were to replan from 𝑠 1 with 𝛿 𝑝𝑟𝑖𝑜𝑟,1 = 0.5, we would now find that 𝑎 2 is the best action, because we find that the set of observations with 50% probability conditioned on o 1 1 and highest reward is…”

Section: Rescaling Of Prior Specializations In Online Planningmentioning

confidence: 99%

“…In the context of the example above, all observations that contain o 1 1 were elements of 𝒰 when planning from 𝑠 0 , so all observations that follow from o 1 1 should be considered as part of 𝒰 1 when planning from 𝑠 1 , resulting in 𝛿 𝑝𝑟𝑖𝑜𝑟,1 = 1.0.…”

Section: Rescaling Of Prior Specializations In Online Planningmentioning

confidence: 99%

“…The optimal offline policy in this example is visualized in Figure 4 The policy in Figure 4-2 is the policy that we wish to execute by performing online planning. For example, we want the first online plan to select 𝑎 0 , then we want the next online plan to select 𝑎 1 1 after o 1 = 0 is received, and for the online plan to select 𝑎 2 1 when o 1 is observed to be 1 or 3.…”

Section: A Motivating Examplementioning

confidence: 99%

See 2 more Smart Citations

Query-driven adaptive sampling

Ayton¹

View full text Add to dashboard Cite

Automated information gathering allows exploration of environments where data is limited and gathering observations introduces risk, such as underwater and planetary exploration. Typically, exploration has been performed in service of a query, with a unique algorithm developed for each mission. Yet this approach does not allow scientists to respond to novel questions as they are raised. In this thesis, we develop a single approach for a broad range of adaptive sampling missions with risk and limited prior knowledge. To achieve this, we present contributions in planning adaptive missions in service of queries, and modeling multi-attribute environments. First, we define a query language suitable for specifying diverse goals in adaptive sampling. The language fully encompasses objectives from previous adaptive sampling approaches, and significantly extends the possible range of objectives. We prove that queries expressible in this language are not biased in a way that avoids information. We then describe a Monte Carlo tree search approach to plan for all queries in our language, using sample based objective estimators embedded within tree search. This approach outperforms methods that maximize information about all variables in hydrocarbon seep search and fire escape scenarios. Next, we show how to plan when the policy must bound risk as a function of reward. By solving approximating problems, we guarantee risk bounds on policies with large numbers of actions and continuous observations, ensuring that risks are only taken when justified by reward. Exploration is limited by the quality of the environment model, so we introduce Gaussian process models with directed acyclic structure to improve model accuracy under limited data. The addition of interpretable structure allows qualitative expert knowledge of the environment to be encoded through structure and parameter constraints. Since expert knowledge may be incomplete, we introduce efficient structure learning over structural models using A* search with bounding conflicts. By placing bounds on likelihood of substructures, we limit the number of structures that are trained, significantly accelerating search. Experiments modeling geographic data show that our model produces more accurate predictions than existing Gaussian process methods, and using bounds allows structure to be learned in 50% of the time.

show abstract