Abstract: A first set of execution platforms is deployed for a set of operations of a training phase of a machine learning model. Prior to the completion of the training phase, a triggering condition for deployment of a different set of execution platforms is detected. The different set of execution platforms is deployed for a subsequent set of training phase operations.

Abstract: Techniques for anomaly detection are described. An exemplary method includes receiving one or more requests to train an anomaly detection machine learning model using feedback-based training, the request to indicate one or more of a type of analysis to perform, a model selection indication, and a configuration for a training dataset; training the anomaly detection machine learning model according to the one or more requests using the training data; performing feedback-based training on the trained anomaly detection machine learning model; and using the retrained anomaly detection machine learning model.

Abstract: Techniques for feedback-based training are described.

Abstract: Techniques are described for automatically scaling (or “auto scaling”) compute resources—for example, virtual machine (VM) instances, containers, or standalone servers—used to support execution of service-oriented software applications and other types of applications that may process heterogeneous workloads. The resource requirements for a software application can be approximated by measuring “worker pool” utilization of instances of each service, where a worker pool represents a number of requests that the service can process concurrently. A scaling service can thus be configured to scale the compute instances provisioned for a service in proportion to worker pool utilization, that is, compute instances can be added as the fleet's worker pools become more “busy,” while compute instances can be removed when worker pools become inactive.

Abstract: A first set of execution platforms is deployed for a set of operations of a training phase of a machine learning model. Prior to the completion of the training phase, a triggering condition for deployment of a different set of execution platforms is detected. The different set of execution platforms is deployed for a subsequent set of training phase operations.

Abstract: A first set of execution platforms is deployed for a set of operations of a training phase of a machine learning model. Prior to the completion of the training phase, a triggering condition for deployment of a different set of execution platforms is detected. The different set of execution platforms is deployed for a subsequent set of training phase operations.

Abstract: Approaches introduce a pre-processing and post-processing framework to a neural network-based approach to identify items represented in an image. For example, a classifier that is trained on several categories can be provided. An image that includes a representation of an item of interest is obtained. Rotated versions of the image are generated and each of a subset of the rotated images is analyzed to determine a probability that a respective image includes an instance of a particular category. The probabilities can be used to determine a probability distribution of output category data, and the data can be analyzed to select an image of the rotated versions of the image. Thereafter, a categorization tree can then be utilized, whereby for the item of interest represented the image, the category of the item can be determined. The determined category can be provided to an item retrieval algorithm to determine primary content for the item of interest.

Abstract: Approaches introduce a pre-processing and post-processing framework to a neural network-based approach to identify items represented in an image. For example, a classifier that is trained on several categories can be provided. An image that includes a representation of an item of interest is obtained. Rotated versions of the image are generated and each of a subset of the rotated images is analyzed to determine a probability that a respective image includes an instance of a particular category. The probabilities can be used to determine a probability distribution of output category data, and the data can be analyzed to select an image of the rotated versions of the image. Thereafter, a categorization tree can then be utilized, whereby for the item of interest represented the image, the category of the item can be determined. The determined category can be provided to an item retrieval algorithm to determine primary content for the item of interest.

Abstract: Embodiments of the subject technology provide for a hybrid OCR approach which combines server and device side processing that can offset disadvantages of performing OCR solely on the server side or the device side. More specifically, the subject technology utilizes image characteristics such as glyph details and image quality measurements to opportunistically schedule OCR processing on the mobile device and/or server. In this regard, text extracted by a “faster” OCR engine (e.g., one with less latency) is displayed to a user, which is then updated by the result of a more accurate OCR engine (e.g., an OCR engine provided by the server). This approach allows factoring in additional parameters such as network latency and user preference for making scheduling decisions. Thus, the subject technology may provide significant gains in terms of reduced latency and increased accuracy by implementing one or more techniques associated with this hybrid OCR approach.

Abstract: Embodiments of the subject technology provide methods and systems of image pre-processing for improving the accuracy of optical character recognition (OCR) and reducing the power consumption on a given computing device (e.g., mobile computing device). The subject technology, in some examples, classifies an image received from a camera of a mobile computing device into one or more classes: 1) normal background, 2) textured background, 3) image with text, 4) image with barcode, 5) image with QR code, and/or 6) image with clutter or “garbage.” Based on the classes associated with the image, the subject technology may forgo certain image processing operations, when the image is not associated with a particular class, in order to save resources (e.g., CPU cycles, battery power, memory usage, etc.) on the mobile computing device.