On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

4 results for “geosteering”

Probabilistic forecasting for geosteering in fluvial successions using a generative adversarial networkphysics.geo-ph

Generative adversarial network improves geosteering in fluvial reservoirs.

problem Improving geosteering in complex reservoirs with high uncertainties.

method Generative adversarial deep neural network (GAN) trained to model fluvial successions.

result Reduces uncertainty and correctly predicts geological features up to 500 meters ahead of drill-bit.

geosteering fluvial reservoirs generative adversarial network deep learning reservoir modeling

2022-07-04abs ↗pdf ↗

Strategic Geosteering Workflow with Uncertainty Quantification and Deep Learning: A Case Study on the Goliat Fieldphysics.geo-ph

Workflow uses deep learning to improve geosteering accuracy in Goliat Field.

problem Uncertainty in geological models and forward simulations affects real-time estimations.

method Offline DNN training, online FlexIES with probabilistic estimation.

result Median probabilistic estimation matches proprietary inversion.

geosteering deep learning Goliat Field uncertainty quantification flexible ensemble smoother

2022-10-27abs ↗pdf ↗

Deep learning for prediction of complex geology ahead of drillingstat.ML

Machine learning improves prediction of complex geology ahead of drilling.

problem Predicting complex geology during drilling in real-time.

method Generative Adversarial Network (GAN) and Forward Deep Neural Network (FDNN) for real-time geological uncertainty reduction.

result Real-time estimates of complex geological uncertainty achieved.

machine learning geosteering geology drilling GAN

2021-04-06abs ↗pdf ↗

Probabilistic model-error assessment of deep learning proxies: an application to real-time inversion of borehole electromagnetic measurementsphysics.geo-ph

Study uses DNNs for real-time EM inversion, highlighting model errors and proposing solutions.

problem Model errors in DNNs affect real-time geosteering decisions in EM measurements.

method Bayesian ensemble smoothing with DNNs for thousands of model evaluations, identifying multimodality.

result Model errors can lead to biased estimates, necessitating error reduction techniques.

Bayesian inversion Deep learning Model errors Real-time geosteering Ensemble smoothing

2022-05-25abs ↗pdf ↗