The technology disclosed relates to classification of process cycle images to predict success or failure of process cycles. The technology disclosed includes capturing and processing images of sections arranged on an image generating chip in genotyping process. Image description features of production cycle images are created and given as input to classifiers. A trained classifier separates successful production images from unsuccessful or failed production images. The failed production images are further classified by a trained root cause classifier into various categories of failure.

Various implementations disclosed herein include devices, systems, and methods that generate floorplans and measurements using a three-dimensional (3D) representation of a physical environment generated based on sensor data.

Machine learning based models recognize objects in images. Specific features of the object are extracted from the image using machine learning based models. The specific features extracted from the image assist deep learning based models in identifying subtypes of a type of object. The system recognizes the objects and collections of objects and determines whether the arrangement of objects violates any predetermined policies. For example, a policy may specify relative positions of different types of objects, height above ground at which certain types of objects are placed, or an expected number of certain types of objects in a collection.

Training deep neural networks requires a large amount of labeled training data. Conventionally, labeled training data is generated by gathering real images that are manually labelled which is very time-consuming. Instead of manually labelling a training dataset, domain randomization technique is used generate training data that is automatically labeled. The generated training data may be used to train neural networks for object detection and segmentation (labelling) tasks. In an embodiment, the generated training data includes synthetic input images generated by rendering three-dimensional (3D) objects of interest in a 3D scene. In an embodiment, the generated training data includes synthetic input images generated by rendering 3D objects of interest on a 2D background image. The 3D objects of interest are objects that a neural network is trained to detect and/or label.

A system and method is provided for automatically creating machine learned fraud detection models. Data received from a plurality of devices can be used to train a model for each of the plurality of entities. Each of the models can be trained using recursive model stacking and each model can output a corresponding score. A second model can be trained for each of the plurality of entities based on the first model and a corresponding output score of the first model. The second model can also be trained using recursive model stacking.

The invention relates to a medical diagnosis assistance system, a medical diagnosis assistance method, and a training method for training an artificial intelligence entity. The medical diagnosis assistance system (100) comprises: an input interface (110) configured to receive medical image data (1) of a patient; a computing device (150) configured to implement: a classification module (151) configured to classify parts of interest, POI (10, 11, 12, 13, 14, 15, 20, 30), comprising objects of interest, OOI, and/or regions of interest, ROI, within the received medical image data (1), and to assign a corresponding reliability metric to each of the classified POI (10, 11, 12, 13, 14, 15, 20, 30); an analysis module (152) configured to determine, based on the POI (10, 11, 12, 13, 14, 15, 20, 30) and the assigned reliability metric, an analysis of the medical image data (1); and an output interface (190) configured to output an output signal (71) indicating the analysis.

A backlight face recognition method, a terminal device using the same, and a computer readable storage medium are provided. The method includes: performing a face detection on each original face image in an original face image sample set to obtain a face frame corresponding to the original face image; capturing the corresponding original face images from the original face image sample set, and obtaining a new face image containing background pixels corresponding to the captured original face images from the original face image sample set; preprocessing all the obtained new face images to obtain a backlight sample set and a normal lighting sample set; and training a convolutional neural network using the backlight sample set and the normal lighting sample set until the convolutional neural network reaches a preset stopping condition. The trained convolutional neural network will improve the accuracy of face recognition in complex background and strong light.

Provided are methods and apparatus for environment-adaptive robotic disinfecting. In an example, provided is a method that can include (i) creating, from digital images, a map of a structure; (ii) identifying a location of a robot in the structure; (iii) segmenting, using a machine learning-based classifying algorithm trained based on object affordance information, the digital images to identify potentially contaminated surfaces within the structure; (iv) creating a map of potentially contaminated surfaces within the structure; (v) calculating a trajectory of movement of the robot to move the robot to a location of a potentially contaminated surface in the potentially contaminated surfaces; and (vi) moving the robot along the trajectory of movement to position a directional decontaminant source adjacent to the potentially contaminated surface. Other methods, systems, and computer-readable media are also disclosed.

A method in an illustrative embodiment includes determining a first set of distilled samples from a first set of samples based on a characteristic distribution of the first set of samples, the first set of samples being associated with a first set of classifications. The method also includes acquiring a first set of characteristic representations associated with the first set of distilled samples. The method also includes adjusting the first set of characteristic representations so that a distance between characteristic representations associated with the same classification is less than a predetermined threshold. The method also includes determining, based on the adjusted first set of characteristic representations, a first set of classification characteristics of the first set of samples and associated with the first set of classifications, the classification characteristics being used to characterize a distribution of characteristic representations of samples having corresponding classifications in the first set of samples.

The present disclosure relates to systems, non-transitory computer-readable media, and methods that implement a pre-defined model container workflow allowing computing devices to flexibly and efficiently define, train, deploy, and maintain machine-learning models. For instance, the disclosed systems can provide scaffolding and boilerplate code for machine-learning models. To illustrate, boilerplate code can include predetermined designs of base classes for common use cases like training, batch inference, etc. In addition, the scaffolding provides an opinionated directory structure for organizing code of a machine-learning model. Further, the disclosed systems can provide containerization and various tooling (e.g., command interface tooling, platform upgrade tooling, and model repository management tooling). Additionally, the disclosed systems can provide out of the box compatibility with one or more different compute instances for increased flexibility and cross-system integration.