An encoder-decoder deep surrogate for reverse time migration in seismic imaging under uncertainty

Freitas, Rodolfo S. M.; Barbosa, Carlos H. S.; Guerra, Gabriel M.; Coutinho, Álvaro L. G. A.; Rochinha, Fernando A.

doi:10.1007/s10596-021-10052-3

Cited by 4 publications

(4 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…DenseED [Freitas et al 2021]. These case studies show how Keras-Prov++ provides new insights into hyperparameter tuning and evidences the relationship of hyperparameter values with metrics such as accuracy.…”

Section: Introductionmentioning

confidence: 91%

“…Our second case study uses the neural network DenseED and its datasets, as proposed by [Freitas et al 2021]. DenseED uses a Physics-guided CNN as a surrogate model to enable the quantification of uncertainties [Zhu and Zabaras 2018].…”

Section: Case Study: Denseedmentioning

confidence: 99%

“…8. As reported in [Freitas et al 2021], the user conducted several hyperparameter fine tunings. Without provenance, it is time consuming to gather the different training executions to compare and register adjustments.…”

Section: Case Study: Denseedmentioning

confidence: 99%

See 2 more Smart Citations

Online Deep Learning Hyperparameter Tuning based on Provenance Analysis

Kunstmann

Pina

Silva

et al. 2021

jidm

View full text Add to dashboard Cite

Training Deep Learning (DL) models require adjusting a series of hyperparameters. Although there are several tools to automatically choose the best hyperparameter configuration, the user is still the main actor to take the final decision. To decide whether the training should continue or try different configurations, the user needs to analyze online the hyperparameters most adequate to the training dataset, observing metrics such as accuracy and loss values. Provenance naturally represents data derivation relationships (i.e., transformations, parameter values, etc.), which provide important support in this data analysis. Most of the existing provenance solutions define their own and proprietary data representations to support DL users in choosing the best hyperparameter configuration, which makes data analysis and interoperability difficult. We present Keras-Prov and its extension, named Keras-Prov++, which provides an analytical dashboard to support online hyperparameter fine-tuning. Different from the current mainstream solutions, Keras-Prov automatically captures the provenance data of DL applications using the W3C PROV recommendation, allowing for hyperparameter online analysis to help the user deciding on changing hyperparameters’ values after observing the performance of the models on a validation set. We provide an experimental evaluation of Keras-Prov++ using AlexNet and a real case study, named DenseED, that acts as a surrogate model for solving equations. During the online analysis, the users identify scenarios that suggest reducing the number of epochs to avoid unnecessary executions and fine-tuning the learning rate to improve the model accuracy.

show abstract

Section: Introductionmentioning

confidence: 91%

Section: Case Study: Denseedmentioning

confidence: 99%

See 1 more Smart Citation

Online Deep Learning Hyperparameter Tuning based on Provenance Analysis

Kunstmann

Pina

Silva

et al. 2021

jidm

View full text Add to dashboard Cite

show abstract

“…The use of data from experiments and simulations to improve the understanding and modeling capabilities of reacting flows has become a new challenge and research opportunity [29,30,31]. Several works have focused on the construction of predictive data-driven machine learning (ML) models able to return accurate predictions at a low cost [32,33,34,35]. Furthermore, datadriven machine learning has been demonstrated to be a reliable tool to build constitutive relations of material properties [36,37].…”

Section: Introductionmentioning

confidence: 99%