Abstract: An image or a spectrum of a surface may be acquired by a computing device, which may be included in a mobile device in some examples. The computing device may extract a measured spectrum from the image and generate a corrected spectrum of the surface. In some examples, the corrected spectrum may be generated to compensate for ambient light influence. The corrected spectrum may be analyzed to provide a result, such as a diagnosis or a product recommendation. In some examples, the result is based, at least in part, on a comparison of the corrected spectrum to reference spectra. In some examples, the result is based, at least in part, on an inference of a machine learning model.

Abstract: Apparatuses and methods can be related to implementing a transformer neural network in a memory. A transformer neural network can be implemented utilizing a resistive memory array. The memory array can comprise programmable memory cells that can be programed and used to store weights of the transformer neural network and perform operations consistent with the transformer neural network.

Abstract: An inspection system for determining wafer defects in semiconductor fabrication may include an image capturing device to capture a wafer image and a classification convolutional neural network (CNN) to determine a classification from a plurality of classes for the captured image. Each of the plurality of classes indicates a type of a defect in the wafer. The system may also include an encoder to encode to convert a training image into a feature vector; a cluster system to cluster the feature vector to generate soft labels for the training image; and a decoder to decode the feature vector into a re-generated image. The system may also include a classification system to determine a classification from the plurality of classes for the training image. The encoder and decoder may be formed from a CNN autoencoder. The classification CNN and the CNN autoencoder may each be a deep neural network.

Abstract: Data may be abstracted and/or masked prior to being provided to a machine learning model for training. A machine learning model may provide a confidence level associated with a result. If the confidence level is too high, the machine learning model or an application including the machine learning model may refrain from providing the result as an output. In some examples, the machine learning model may provide a “second best” result that has an acceptable confidence level. In other examples, an error signal may be provided as the output. In accordance with examples of the present disclosure, data may be abstracted and/or masked prior to being provided to a machine learning model for training and confidence levels of results of the trained machine learning model may be used to determine when a result should be withheld.

Abstract: A system for generating simulated data is disclosed. The system may determine items of content utilized by a network. The system may also retrieve one or more data patterns associated with one or more features associated with the content. The system may also determine a plurality of indices associated with the data patterns. The system may also generate, based on the data patterns and the plurality of indices, simulated data associated with the content.

Abstract: Apparatuses and methods can be related to implementing a Bayesian neural network in a memory. A Bayesian neural network can be implemented utilizing a resistive memory array. The memory array can comprise programmable memory cells that can be programed and used to store weights of the Bayesian neural network and perform operations consistent with the Bayesian neural network.

Abstract: Embodiments of the disclosure are drawn to projecting light on a surface and analyzing the scattered light to obtain spatial information of the surface and generate a three dimensional model of the surface. The three dimensional model may then be analyzed to calculate one or more surface characteristics, such as roughness. The surface characteristics may then be analyzed to provide a result, such as a diagnosis or a product recommendation. In some examples, a mobile device is used to analyze the surface.

Abstract: Apparatuses and methods can be related to implementing a binary neural network in memory. A binary neural network can be implemented utilizing a resistive memory array. The memory array can comprise programmable memory cells that can be programed and used to store weights of the binary neural network and perform operations consistent with the binary neural network. The weights of the binary neural network can correspond to non-zero values.

Abstract: Methods, apparatuses, and systems associated with anomaly detection and resolution are described. Examples can include detecting, via a sensor of a robot, an object in a path of the robot while the robot is performing a task in an environment and classifying the object as an anomaly or a non-anomaly and the environment as anomalous or non-anomalous using a machine learning model. Examples can include proceeding with the task responsive to classification of the object as a non-anomaly and the environment as non-anomalous and resolving the anomaly or the anomalous environment and proceeding with the task responsive to classification of the object as an anomaly or the environment as anomalous.

Abstract: An image or a spectrum of a surface may be acquired by a computing device, which may be included in a mobile device in some examples. The computing device may extract a measured spectrum from the image and generate a corrected spectrum of the surface. In some examples, the corrected spectrum may be generated to compensate for ambient light influence. The corrected spectrum may be analyzed to provide a result, such as a diagnosis or a product recommendation. In some examples, the result is based, at least in part, on a comparison of the corrected spectrum to reference spectra. In some examples, the result is based, at least in part, on an inference of a machine learning model.

Abstract: Apparatuses and methods can be related to implementing a Bayesian neural network in a memory. A Bayesian neural network can be implemented utilizing a resistive memory array. The memory array can comprise programmable memory cells that can be programed and used to store weights of the Bayesian neural network and perform operations consistent with the Bayesian neural network.

Abstract: Apparatuses and methods can be related to implementing a transformer neural network in a memory. A transformer neural network can be implemented utilizing a resistive memory array. The memory array can comprise programmable memory cells that can be programed and used to store weights of the transformer neural network and perform operations consistent with the transformer neural network.

Abstract: Embodiments of the disclosure are drawn to projecting light on a surface and analyzing the scattered light to obtain spatial information of the surface and generate a three dimensional model of the surface. The three dimensional model may then be analyzed to calculate one or more surface characteristics, such as roughness. The surface characteristics may then be analyzed to provide a result, such as a diagnosis or a product recommendation. In some examples, a mobile device is used to analyze the surface.

Abstract: Data may be abstracted and/or masked prior to being provided to a machine learning model for training. A machine learning model may provide a confidence level associated with a result. If the confidence level is too high, the machine learning model or an application including the machine learning model may refrain from providing the result as an output. In some examples, the machine learning model may provide a “second best” result that has an acceptable confidence level. In other examples, an error signal may be provided as the output. In accordance with examples of the present disclosure, data may be abstracted and/or masked prior to being provided to a machine learning model for training and confidence levels of results of the trained machine learning model may be used to determine when a result should be withheld.

Abstract: An inspection system for determining wafer defects in semiconductor fabrication may include an image capturing device to capture a wafer image and a classification convolutional neural network (CNN) to determine a classification from a plurality of classes for the captured image. Each of the plurality of classes indicates a type of a defect in the wafer. The system may also include an encoder to encode to convert a training image into a feature vector; a cluster system to cluster the feature vector to generate soft labels for the training image; and a decoder to decode the feature vector into a re-generated image. The system may also include a classification system to determine a classification from the plurality of classes for the training image. The encoder and decoder may he formed from a CNN autoencoder. The classification CNN and the CNN autoencoder may each be a deep neural network.

Abstract: In the methods and systems, optical images of inspection care areas on a semiconductor wafer are acquired and analyzed to detect defects. However, during this analysis, the same threshold setting is not used for all inspection care areas. Instead, care areas are grouped into different care area groups, based on different design layouts and properties. Each group is associated with a corresponding threshold setting that is optimal for detecting defects in the inspection care areas belonging to that group. The assignment of the care areas to the different groups and the association of the different threshold settings with the different groups are noted in an index. This index is accessible during the analysis and used to ensure that each of the inspection care areas in a specific care area group is analyzed based on a corresponding threshold setting that is optimal for that specific care area group.

Abstract: Systems, methods, and computer program products for design rules checking in which the waiver of design rules is optimized while ensuring compliant designs that are manufacturable. A first design rule and a plurality of patterns of a layout that violate the first design rule are received by a design rule waiver system. The design rule waiver system may process the first design rule to extract a plurality of descriptors that can be perturbed. The design rule waiver system may perturb an attribute associated with at least one of the plurality of descriptors extracted from the first design rule in order to define a second design rule that is satisfied by the plurality of patterns.

Abstract: Systems, methods, and computer program products for design rules checking in which the waiver of design rules is optimized while ensuring compliant designs that are manufacturable. A first design rule and a plurality of patterns of a layout that violate the first design rule are received by a design rule waiver system. The design rule waiver system may process the first design rule to extract a plurality of descriptors that can be perturbed. The design rule waiver system may perturb an attribute associated with at least one of the plurality of descriptors extracted from the first design rule in order to define a second design rule that is satisfied by the plurality of patterns.

Abstract: In the methods and systems, optical images of inspection care areas on a semiconductor wafer are acquired and analyzed to detect defects. However, during this analysis, the same threshold setting is not used for all inspection care areas. Instead, care areas are grouped into different care area groups, based on different design layouts and properties. Each group is associated with a corresponding threshold setting that is optimal for detecting defects in the inspection care areas belonging to that group. The assignment of the care areas to the different groups and the association of the different threshold settings with the different groups are noted in an index. This index is accessible during the analysis and used to ensure that each of the inspection care areas in a specific care area group is analyzed based on a corresponding threshold setting that is optimal for that specific care area group.

Abstract: Candidate layout patterns can be generated using a generative model trained based on known data, such as historical hot spot data, features extraction, and geometrical primitives. The generative model can be sampled to obtain candidate layouts that can be ranked and repaired using error optimization, design rule checking, optical proximity checking, and other methods to ensure that resulting candidates are manufacturable.