| 1. |
|
|
| 2. |
- Abbaszadeh Shahri, A., et al.
(författare)
-
A Novel Approach to Uncertainty Quantification in Groundwater Table Modeling by Automated Predictive Deep Learning
- 2022
-
Ingår i: Natural Resources Research. - : Springer Nature. - 1520-7439 .- 1573-8981. ; 31:3, s. 1351-1373
-
Tidskriftsartikel (refereegranskat)abstract
- Uncertainty quantification (UQ) is an important benchmark to assess the performance of artificial intelligence (AI) and particularly deep learning ensembled-based models. However, the ability for UQ using current AI-based methods is not only limited in terms of computational resources but it also requires changes to topology and optimization processes, as well as multiple performances to monitor model instabilities. From both geo-engineering and societal perspectives, a predictive groundwater table (GWT) model presents an important challenge, where a lack of UQ limits the validity of findings and may undermine science-based decisions. To overcome and address these limitations, a novel ensemble, an automated random deactivating connective weights approach (ARDCW), is presented and applied to retrieved geographical locations of GWT data from a geo-engineering project in Stockholm, Sweden. In this approach, the UQ was achieved via a combination of several derived ensembles from a fixed optimum topology subjected to randomly switched off weights, which allow predictability with one forward pass. The process was developed and programmed to provide trackable performance in a specific task and access to a wide variety of different internal characteristics and libraries. A comparison of performance with Monte Carlo dropout and quantile regression using computer vision and control task metrics showed significant progress in the ARDCW. This approach does not require changes in the optimization process and can be applied to already trained topologies in a way that outperforms other models.
|
|
| 3. |
- Abbaszadeh Shahri, A., et al.
(författare)
-
Artificial intelligence models to generate visualized bedrock level : a case study in Sweden
- 2020
-
Ingår i: Modeling Earth Systems and Environment. - : Springer. - 2363-6203 .- 2363-6211. ; 6:3, s. 1509-1528
-
Tidskriftsartikel (refereegranskat)abstract
- Assessment of the spatial distribution of bedrock level (BL) as the lower boundary of soil layers is associated with many uncertainties. Increasing our knowledge about the spatial variability of BL through high resolution and more accurate predictive models is an important challenge for the design of safe and economical geostructures. In this paper, the efficiency and predictability of different artificial intelligence (AI)-based models in generating improved 3D spatial distributions of the BL for an area in Stockholm, Sweden, were explored. Multilayer percepterons, generalized feed-forward neural network (GFFN), radial based function, and support vector regression (SVR) were developed and compared to ordinary kriging geostatistical technique. Analysis of the improvement in progress using confusion matrixes showed that the GFFN and SVR provided closer results to realities. The ranking of performance accuracy using different statistical errors and precision/recall curves also demonstrated the superiority and robustness of the GFFN and SVR compared to the other models. The results indicated that in the absence of measured data the AI models are flexible and efficient tools in creating more accurate spatial 3D models. Analyses of confidence intervals and prediction intervals confirmed that the developed AI models can overcome the associated uncertainties and provide appropriate prediction at any point in the subsurface of the study area.
|
|
| 4. |
|
|
| 5. |
|
|
