Microsoft Technology Licensing, LLC

Amazon Technologies, Inc.

Barath Balasubramanian

Olivier Colle

Evan Bjorn-Thomas Cohen

Ranju Das

Prakash Krishnan

Joaquin Zepeda Salvatierra

Niels Brouwers

Anant Patel

Shaun Ryan James McDowell

William James Staples

Rahul Bhotika

Gregory Scott Lindhorst

Avinash Aghoram Ravichandran

Gurumurthy Swaminathan

Rakesh Madhavan Nambiar

Alexander Jon Dobin

Evan Cohen

Radu Bogdan Gruian

Ashwin Sathya Raghunathan

Irina Koulinitch

Muralidhar Sathsahayaraman

Daniel Joseph Kappes

Mohammed Amine Benmouffok

Unnati J Dani

Jade D Naaman

Shrey Sangal

Anushri Mainthia

Shaun Ryan James Mcdowell

Techniques for feedback-based training may include selecting a scoring machine learning model based at least in part on a test metric, and applying the model on an unlabeled dataset to generate, per dataset item of the unlabeled dataset, a prediction and an importance ranking score for the prediction. Techniques for feedback-based training may further include selecting, based on the importance ranking scores, a result of the application of the model on the unlabeled dataset, providing the result and requesting feedback on the result via a graphical user interface, receiving the feedback via the graphical user interface, adding data from the unlabeled dataset into a training dataset when the feedback indicates a verified result, and retraining the model using the training dataset with the data added from the unlabeled dataset to generate a retrained model.

Feedback-based training for anomaly detection

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.

Anomaly detection using feedback training

Techniques for anomaly detection are described. An exemplary method includes receiving a request to create a training data set from at least one image, the request to include an indication of the at least one image and at least one indication of an operation to perform on the at least one image to generate a plurality of images from the at least one image; creating a training dataset by extracting one or more chunks from a first at least one image according to the request; and receiving one or more requests to train an anomaly detection machine learning model using the created training dataset; and training an anomaly detection machine learning model according to one or more requests using the created training data.

Generation of training data from a single image

Examples of the disclosure provide for modifying an application from a mobile device. An instruction to modify an application including one or more documents is received at the mobile device. It is determined at the mobile device whether the received instruction is associated with a first type of modification and/or a second type of modification. In response to determining that the received instruction is associated with the first type of modification, a local instance of the document stored at the mobile device is modified at the mobile device. In response to determining that the received instruction is associated with the second type of modification, a remote instance of the document stored at a server device is modified at the server device. Aspects of the disclosure enable a mobile device to strategically modify an application to facilitate managing local resources.

Creating and modifying applications from a mobile device

A workflow development system is described that enables users to easily develop and deploy bots. The system provides a graphical user interface (GUI) that enables a user to select a plurality of steps for inclusion in a workflow and within which each user-selected step is represented as an interactive object. The plurality of steps includes at least one step of receiving a message directed to a bot or sending a message on behalf of the bot and at least one step of invoking an application to perform an action in response to receiving the message directed to the bot or to receive information upon which the sending of the message on behalf of the bot will be based. The system also receives information about each of the user-selected steps included in the workflow and generates workflow logic based on the received information that is executable to run the workflow.

Bot creation with workflow development system

Examples of the disclosure provide for receiving a template at a design surface, the template representing a visual layout, logic, and data schema for an application. Received source data is bound to the template based on the data schema. At least one data storage location is selected from identified data storage locations. Based on the selection of the one data storage location, at least a portion of the source data is extracted and sent to the data storage location for storage. The extracted data is rewired to bind the extracted data at the data storage location to the template such that access of the application generated using the template provides access to the extracted data at the data storage location.

Generation of an application from template

Examples of the disclosure provide for receiving a data set at a design surface from a data source, and inferring a first data type from the data set. A first control type is generated for the inferred first data type. A second data type is inferred from the data set and a second control type is generated for the inferred second data type, with the first control type and the second control type being different. A layout is generated for an application document that includes the first control type and the second control type.

Generation of an application from data

Growing community of inventors

Bothell, WA, United States of America

Barath Balasubramanian