Abstract: An active learning framework is provided that employs a plurality of machine learning components that operate over iterations of a training phase followed by an active learning phase. In each iteration of the training phase, the machine learning components are trained from a pool of labeled observations. In the active learning phase, the machine learning components are configured to generate metrics used to control sampling of unlabeled observations for labeling such that newly labeled observations are added to a pool of labeled observations for the next iteration of the training phase. The machine learning components can include an inspection (or primary) learning component that generates a predicted label and uncertainty score for an unlabeled observation, and at least one additional component that generates a quality metric related to the unlabeled observation or the predicted label. The uncertainty score and quality metric(s) can be combined for efficient sampling of observations for labeling.

Abstract: A method for controlling a subsea vehicle. The method includes receiving sensor data representing a subsea environment from one or more sensors of the subsea vehicle. The method identifies one or more objects present in the subsea environment based on the sensor data using an artificial intelligence machine. The method transmits at least a portion of the sensor data, including an identification of the one or more objects, to a user interface. The method includes receiving a requested vehicle task from the user interface. The requested vehicle task being selected by a user via the user interface. The method performs the requested vehicle task without vehicle position control from the user.

Abstract: Methods, computing systems, and computer-readable media for interpreting seismic data, of which the method includes receiving seismic data representing a subterranean volume, and determining a feature-likelihood attribute of at least a portion of a section of the seismic data. The feature-likelihood attribute comprises a value for elements of the section, the value being based on a likelihood that the element represents part of a subterranean feature. The method also includes identifying contours of the subterranean feature based in part on the feature-likelihood attribute of the section, and determining a polygonal line that approximates the subterranean feature.

Abstract: A method for detecting an unknown fault in a target seismic volume. The method includes generating a number of patches from a training seismic volume that is separate from the target seismic volume, where a patch includes a set of training areas, generating a label for assigning to the patch, where the label represents a subset, of the set of training areas, intersected by an known fault specified by a user in the training seismic volume, training, during a training phase and based at least on the label and the training seismic volume, a machine learning model, and generating, by applying the machine learning model to the target seismic volume during a prediction phase subsequent to the training phase, a result to identify the unknown fault in the target seismic volume.

Abstract: A method can include receiving labeled images; acquiring unlabeled images; performing active learning by training an inspection learner using at least a portion of the labeled images to generate a trained inspection learner that outputs information responsive to receipt of one of the unlabeled images by the trained inspection learner; based at least in part on the information, making a decision to call for labeling of the one of the unlabeled images; receiving a label for the one of the unlabeled images; and further training the inspection learner using the label.

Abstract: Systems, computer-readable media, and methods for generating machine learning training data by obtaining reservoir data, determining subsections of the reservoir data, labeling the subsections of the reservoir data to generate labeled reservoir data, and feeding the labeled reservoir data into an artificial neural network. The reservoir data can be labeled using analysis data or based on interpretive input from an interpreter.

Abstract: A method can include receiving seismic image data; processing the received seismic image data to generate stratigraphic information using a trained convolution neural network that includes channels subjected to convolution, activation and pooling that reduce spatial resolution and subjected to deconvolution and concatenation that increase spatial resolution; and enhancing the seismic image data using the stratigraphic information to generate an enhanced seismic image.

Abstract: A method can include selecting a type of geophysical data; selecting a type of algorithm; generating synthetic geophysical data based at least in part on the algorithm; training a deep learning framework based at least in part on the synthetic geophysical data to generate a trained deep learning framework; receiving acquired geophysical data for a geologic environment; implementing the trained deep learning framework to generate interpretation results for the acquired geophysical data; and outputting the interpretation results.

Abstract: A method is disclosed and includes receiving a seismic cube. The seismic cube includes a three-dimensional image of a portion of a subsurface area. The method further includes providing the seismic cube to a machine learning process. The machine learning process includes one or more neural networks used for predicting a location of a subsurface seismic layer in the received seismic cube. The method also includes receiving, from the machine learning process, the prediction of the location of the subsurface seismic layer in the seismic cube.

Abstract: An active learning framework is provided that employs a plurality of machine learning components that operate over iterations of a training phase followed by an active learning phase. In each iteration of the training phase, the machine learning components are trained from a pool of labeled observations. In the active learning phase, the machine learning components are configured to generate metrics used to control sampling of unlabeled observations for labeling such that newly labeled observations are added to a pool of labeled observations for the next iteration of the training phase. The machine learning components can include an inspection (or primary) learning component that generates a predicted label and uncertainty score for an unlabeled observation, and at least one additional component that generates a quality metric related to the unlabeled observation or the predicted label. The uncertainty score and quality metric(s) can be combined for efficient sampling of observations for labeling.

Abstract: A method can include receiving seismic image data of a geologic region and interpretation information of the seismic image data for a geologic feature in the geologic region, where the seismic image data include a geologic feature class imbalance; shifting the geologic feature class imbalance toward a class of the geologic feature by increasing the spatial presence of the geologic feature in the seismic image data to generate training data; and training a neural network using the training data to generate a trained neural network.

Abstract: Methods, computing systems, and computer-readable media for interpreting seismic data, of which the method includes receiving seismic data representing a subterranean volume, and determining a feature-likelihood attribute of at least a portion of a section of the seismic data. The feature-likelihood attribute comprises a value for elements of the section, the value being based on a likelihood that the element represents part of a subterranean feature. The method also includes identifying contours of the subterranean feature based in part on the feature-likelihood attribute of the section, and determining a polygonal line that approximates the subterranean feature.

Abstract: A method for controlling a subsea vehicle. The method includes receiving sensor data representing a subsea environment from one or more sensors of the subsea vehicle. The method identifies one or more objects present in the subsea environment based on the sensor data using an artificial intelligence machine. The method transmits at least a portion of the sensor data, including an identification of the one or more objects, to a user interface. The method includes receiving a requested vehicle task from the user interface. The requested vehicle task being selected by a user via the user interface. The method performs the requested vehicle task without vehicle position control from the user.

Abstract: A method can include receiving seismic image data; processing the received seismic image data to generate stratigraphic information using a trained convolution neural network that includes channels subjected to convolution, activation and pooling that reduce spatial resolution and subjected to deconvolution and concatenation that increase spatial resolution; and enhancing the seismic image data using the stratigraphic information to generate an enhanced seismic image.

Abstract: A method can include receiving seismic image data of a geologic region and interpretation information of the seismic image data for a geologic feature in the geologic region, where the seismic image data include a geologic feature class imbalance; shifting the geologic feature class imbalance toward a class of the geologic feature by increasing the spatial presence of the geologic feature in the seismic image data to generate training data; and training a neural network using the training data to generate a trained neural network.

Abstract: Systems, methods, and non-transitory computer-readable media for processing geological data, of which the method includes receiving a geological data processing tool at a client system. The geological data processing tool includes artificial intelligence, and the geological data processing tool is generated by a geological processing tool provider. The method also includes obtaining training data for the geological data processing tool. The training data includes a plurality of labels. The method also includes training the geological data processing tool based on the training data, receiving data representing a physical, subterranean volume, identifying one or more geological features in the subterranean volume by using the geological data processing tool after training the geological data processing tool, and modifying, using the client system, one or more parameters of the geological data processing tool, or one or more labels of the plurality of labels.

Abstract: A method can include selecting a type of geophysical data; selecting a type of algorithm; generating synthetic geophysical data based at least in part on the algorithm; training a deep learning framework based at least in part on the synthetic geophysical data to generate a trained deep learning framework; receiving acquired geophysical data for a geologic environment; implementing the trained deep learning framework to generate interpretation results for the acquired geophysical data; and outputting the interpretation results.

Abstract: A method is disclosed and includes receiving a seismic cube. The seismic cube includes a three-dimensional image of a portion of a subsurface area. The method further includes providing the seismic cube to a machine learning process. The machine learning process includes one or more neural networks used for predicting a location of a subsurface seismic layer in the received seismic cube. The method also includes receiving, from the machine learning process, the prediction of the location of the subsurface seismic layer in the seismic cube.

Abstract: A method can include receiving points representative of at least a portion of a surface of a multi-dimensional geobody; partitioning the points; computing smooth compactly supported basis functions based at least in part on differential surface areas associated with the partitioning of the points; approximating an indicator function for a body based at least in part on the computed basis functions; and, based at least in part on values of the approximated indicator function, generating a mesh that represents a surface of the body.

Abstract: Systems, computer-readable media, and methods for generating machine learning training data by obtaining reservoir data, determining subsections of the reservoir data, labeling the subsections of the reservoir data to generate labeled reservoir data, and feeding the labeled reservoir data into an artificial neural network. The reservoir data can be labeled using analysis data or based on interpretive input from an interpreter.