Google Inc.

Servicenow Canada Inc.

Philippe J Beaudoin

Shibl Mourad

Caitlin Kelly Phillips

Marc-Antoine Courteau

Steven Chi Truong

Manas Tungare

Bernhard Bauer

Chris Bentzel

Sherif Elsayed-Ali

Systems and methods relating to machine learning. An edge device runs a new data point on a first neural network and determines activations on the layers within that neural network. The first neural network is a fully trained network based on a second neural network on a server. The activation data for the various layers in the first neural network are, starting with the output layer, sequentially transmitted to the server. The server continuously receives this activation data and continuously compares it with previously encountered activation data for the second neural network. If the received activation data is within an expected range, then the edge device is instructed to stop sending activation data. Otherwise, the server continues to receive the activation data for the other layers until the new data point is received by the server or the activation data is within the expected range of previously encountered activation data.

Data point suitability determination from edge device neural networks

A method for designing an image filter, the method being executed by a machine learning algorithm (MLA), the method comprising: receiving unfiltered images, a desirable visual task and an undesirable visual task; for each unfiltered image, receiving a first label indicative of whether the desirable visual task is accomplishable and a second label indicative of whether the undesirable visual task is accomplishable; filtering the unfiltered image using a virtual filter model representative of the image filter; using the filtered images, training the MLA to perform the desirable visual task and prevent the undesirable visual task to be performed; using test images, determining a first efficiency for the MLA to perform the desirable task and a second efficiency for the MLA to prevent the undesirable task; adjusting a parameter of the virtual filter model based on the first and second efficiencies; and outputting the adjusted parameter.

Method and system for designing an optical filter

Systems and methods relating to machine learning. A machine learning model is trained on a data processing system with the most amount of resources available to it. The resulting trained model is the basis for various stripped down versions that are suitable for execution on systems with less resources available. Specific system/version combinations are tested to determine which combination works best with specific systems and versions. To update installed trained models, a differential compression method can be used so that only an update difference needs to be uploaded to the system being updated.

Machine learning model hardware configuration based optimization

Activating a private browsing mode for a browser can include receiving an electronic document within the browser and detecting an indicator associated with the electronic document using a processor, wherein the indicator is correlated with the private browsing mode of the browser. Responsive to detecting the indicator, the electronic document can be rendered within a view of the browser in which private browsing mode is activated.

Invoking a private browsing mode by selection of a visual control element within a browser tab

A computer-implemented method for an anonymous personalized recommendation is provided. The method includes receiving select server fingerprints selected from server fingerprints based on predetermined metrics. The server fingerprints correspond to a plurality of public entities and each server fingerprint includes encoded information about a corresponding public entity. The method also includes generating a user fingerprint based on user information associated with a user, the user fingerprint comprising encoded user information. The method further includes comparing the user fingerprint with the select server fingerprints to select a server fingerprint for recommendation, generating a recommendation of a public entity corresponding to the server fingerprint for recommendation, and displaying the recommendation. Systems and machine-readable media are also provided.

Anonymous personalized recommendation method

A computer-implemented method for an anonymous personalized recommendation is provided. The method includes receiving select server fingerprints selected from server fingerprints based on predetermined metrics. The server fingerprints correspond to a plurality of public entities and each server fingerprint includes encoded information about a corresponding public entity. The method also includes generating a user fingerprint based on user information associated with a user, the user fingerprint comprising encoded user information. The method further includes comparing the user fingerprint with the select server fingerprints to select a server fingerprint for recommendation, and generating a recommendation of a public entity corresponding to the server fingerprint for recommendation. Systems and machine-readable media are also provided.

Growing community of inventors

Montreal, Canada

Philippe J Beaudoin