Beyond Local Nash Equilibria for Adversarial Networks

Oliehoek, Frans A.; Savani, Rahul; Gállego, José Ramón; Pol, Elise van der; Groß, Roderich

doi:10.1007/978-3-030-31978-6_7

Cited by 20 publications

(12 citation statements)

References 22 publications

(34 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where x refers to the training data sampled from the data distribution p data (x), and z to a noise variable sampled from some prior p(z). The discriminator is thus trained to maximise V (D), while the generator aims at minimising V (D); the two networks play a minimax game until a Nash equilibrium is (hopefully) reached (Goodfellow et al, 2014;Che et al, 2016;Oliehoek et al, 2018).…”

Section: Wasserstein Generative Adversarial Network -Gradient Penaltmentioning

confidence: 99%

Accelerating Bayesian microseismic event location with deep learning

Mancini¹,

Piras²,

Ferreira³

et al. 2021

Preprint

View full text Add to dashboard Cite

Abstract. We present a series of new open source deep learning algorithms to accelerate Bayesian full waveform point source inversion of microseismic events. Inferring the joint posterior probability distribution of moment tensor components and source location is key for rigorous uncertainty quantification. However, the inference process requires forward modelling of micro-seismic traces for each set of parameters explored by the sampling algorithm, which makes the inference very computationally intensive. In this paper we focus on accelerating this process by training deep learning models to learn the mapping between source location and seismic traces, for a given 3D heterogeneous velocity model, and a fixed isotropic moment tensor for the sources. These trained emulators replace the expensive solution of the elastic wave equation in the inference process. We compare our results with a previous study that used emulators based on Gaussian Processes to invert microseismic events. For fairness of comparison, we train our emulators on the same microseismic traces and using the same geophysical setting. We show that all of our models provide more accurate predictions and ∼100 times faster predictions than the method based on Gaussian Processes, and a O(105) speed-up factor over a pseudo-spectral method for waveform generation. For example, a 2-s long synthetic trace can be generated in ∼10 ms on a common laptop processor, instead of ∼1 hr using a pseudo-spectral method on a high-profile Graphics Processing Units card. We also show that our inference results are in excellent agreement with those obtained from traditional location methods based on travel time estimates. The speed, accuracy and scalability of our open source deep learning models pave the way for extensions of these emulators to generic source mechanisms and application to joint Bayesian inversion of moment tensor components and source location using full waveforms.

show abstract

Section: Wasserstein Generative Adversarial Network -Gradient Penaltmentioning

confidence: 99%

Accelerating Bayesian microseismic event location with deep learning

Mancini¹,

Piras²,

Ferreira³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…More formally, we can define a value function as follows: where x refers to the training data sampled from the data distribution p data (x), and z to a noise variable sampled from some prior p(z). The discriminator is thus trained to maximize V (D), while the generator aims at minimizing V (D); the two networks play a minimax game until a Nash equilibrium is (hopefully) reached (Goodfellow et al, 2014;Che et al, 2016;Oliehoek et al, 2018).…”

Section: Wasserstein Generative Adversarialmentioning

confidence: 99%

Accelerating Bayesian microseismic event location with deep learning

Mancini

Piras²,

Ferreira³

et al. 2021

Solid Earth

View full text Add to dashboard Cite

Abstract. We present a series of new open-source deep-learning algorithms to accelerate Bayesian full-waveform point source inversion of microseismic events. Inferring the joint posterior probability distribution of moment tensor components and source location is key for rigorous uncertainty quantification. However, the inference process requires forward modelling of microseismic traces for each set of parameters explored by the sampling algorithm, which makes the inference very computationally intensive. In this paper we focus on accelerating this process by training deep-learning models to learn the mapping between source location and seismic traces for a given 3D heterogeneous velocity model and a fixed isotropic moment tensor for the sources. These trained emulators replace the expensive solution of the elastic wave equation in the inference process. We compare our results with a previous study that used emulators based on Gaussian processes to invert microseismic events. For fairness of comparison, we train our emulators on the same microseismic traces and using the same geophysical setting. We show that all of our models provide more accurate predictions, ∼ 100 times faster predictions than the method based on Gaussian processes, and a 𝒪(105) speed-up factor over a pseudo-spectral method for waveform generation. For example, a 2 s long synthetic trace can be generated in ∼ 10 ms on a common laptop processor, instead of ∼ 1 h using a pseudo-spectral method on a high-profile graphics processing unit card. We also show that our inference results are in excellent agreement with those obtained from traditional location methods based on travel time estimates. The speed, accuracy, and scalability of our open-source deep-learning models pave the way for extensions of these emulators to generic source mechanisms and application to joint Bayesian inversion of moment tensor components and source location using full waveforms.

show abstract

“…During training, both networks engage in competition until a Nash equilibrium is reached. In game theory, Nash equilibrium is a status where no player is able improve or deviate his payoff [40].…”

Section: The Generator and The Discriminatormentioning

confidence: 99%

Data-Efficient Domain Adaptation for Semantic Segmentation of Aerial Imagery Using Generative Adversarial Networks

et al. 2020

View full text Add to dashboard Cite

Despite the significant advances noted in semantic segmentation of aerial imagery, a considerable limitation is blocking its adoption in real cases. If we test a segmentation model on a new area that is not included in its initial training set, accuracy will decrease remarkably. This is caused by the domain shift between the new targeted domain and the source domain used to train the model. In this paper, we addressed this challenge and proposed a new algorithm that uses Generative Adversarial Networks (GAN) architecture to minimize the domain shift and increase the ability of the model to work on new targeted domains. The proposed GAN architecture contains two GAN networks. The first GAN network converts the chosen image from the target domain into a semantic label. The second GAN network converts this generated semantic label into an image that belongs to the source domain but conserves the semantic map of the target image. This resulting image will be used by the semantic segmentation model to generate a better semantic label of the first chosen image. Our algorithm is tested on the ISPRS semantic segmentation dataset and improved the global accuracy by a margin up to 24% when passing from Potsdam domain to Vaihingen domain. This margin can be increased by addition of other labeled data from the target domain. To minimize the cost of supervision in the translation process, we proposed a methodology to use these labeled data efficiently.

show abstract

Beyond Local Nash Equilibria for Adversarial Networks

Cited by 20 publications

References 22 publications

Accelerating Bayesian microseismic event location with deep learning

Accelerating Bayesian microseismic event location with deep learning

Accelerating Bayesian microseismic event location with deep learning

Data-Efficient Domain Adaptation for Semantic Segmentation of Aerial Imagery Using Generative Adversarial Networks

Contact Info

Product

Resources

About