Abstract: Apparatuses, systems, and techniques to facilitate feature detection of a manufactured object such as a PCB using combined images of said manufactured object. In at least one embodiment, an automated optical inspection system (AOI) comprising one or more neural networks can infer based, at least in part, on combined images of a PCB the existence of defects on said PCB.

Abstract: Apparatuses, systems, and techniques to train neural networks to perform classification. In at least one embodiment, one or more neural networks are trained to perform classification based, at least in part, on grouping one or more sets of neural network training data according to behaviors of one or more objects within one or more images represented by the training data.

Abstract: Personalized coaching is provided to users of an application, such as players of an electronic gaming application. Data can be obtained that demonstrates how skilled users utilize an application, such as how professional players play a game. This data can be used to train a machine learning model for the game. Gameplay data for an identified player can be obtained, and related information provided as input to the trained model. The model can infer one or more actions or strategies to be taken by the player in order to achieve a determined goal. The information can be conveyed to the player using visual, audio, or haptic guidance during gameplay, or can be provided offline, such as with video or rendered replay of the game session. The types of advice or coaching given can vary depending upon factors such as the goals, skill level, and preferences of the player, and can update over time.

Abstract: Personalized coaching is provided to users of an application, such as players of an electronic gaming application. Data can be obtained that demonstrates how skilled users utilize an application, such as how professional players play a game. This data can be used to train a machine learning model for the game. Gameplay data for an identified player can be obtained, and related information provided as input to the trained model. The model can infer one or more actions or strategies to be taken by the player in order to achieve a determined goal. The information can be conveyed to the player using visual, audio, or haptic guidance during gameplay, or can be provided offline, such as with video or rendered replay of the game session. The types of advice or coaching given can vary depending upon factors such as the goals, skill level, and preferences of the player, and can update over time.

Abstract: Personalized coaching is provided to users of an application, such as players of an electronic gaming application. Data can be obtained that demonstrates how skilled users utilize an application, such as how professional players play a game. This data can be used to train a machine learning model for the game. Gameplay data for an identified player can be obtained, and related information provided as input to the trained model. The model can infer one or more actions or strategies to be taken by the player in order to achieve a determined goal. The information can be conveyed to the player using visual, audio, or haptic guidance during gameplay, or can be provided offline, such as with video or rendered replay of the game session. The types of advice or coaching given can vary depending upon factors such as the goals, skill level, and preferences of the player, and can update over time.

Abstract: Systems and methods for contextual and cross application threat detection in cloud applications in accordance with embodiments of the invention are disclosed. In one embodiment, a method for detecting threat activity in a cloud application using past activity data from cloud applications includes receiving activity data concerning actions performed by a user account associated with a user within a monitored cloud application, receiving external contextual data about the user that does not concern actions performed using the user account within the monitored cloud application, where the external contextual data is retrieved from outside of the monitored cloud application, deriving a baseline user profile using the activity data and external contextual data and associating the baseline user profile with the user account, and determining the likelihood of anomalous activity using the baseline user profile.

Abstract: Systems and methods for contextual and cross application threat detection in cloud applications in accordance with embodiments of the invention are disclosed. In one embodiment, a method for detecting threat activity in a cloud application using past activity data from cloud applications includes receiving activity data concerning actions performed by a user account associated with a user within a monitored cloud application, receiving external contextual data about the user that does not concern actions performed using the user account within the monitored cloud application, where the external contextual data is retrieved from outside of the monitored cloud application, deriving a baseline user profile using the activity data and external contextual data and associating the baseline user profile with the user account, and determining the likelihood of anomalous activity using the baseline user profile.