Zestfinance, Inc.

Microsoft Technology Licensing, LLC

John Wickens Lamb Merrill

Jerome Louis Budzik

Sean Javad Kamkar

Mark Frederick Eberstein

Jose Efrain Valentin

Feng Li

John Joseph Beahan, Jr

Michael Egan Van Veen

Mark Jeffrey Weinberg

Tandy W Trower, Ii

John Candido

Douglas C Merrill

Yachen Yan

Michael Edward Ruberry

Benjamin Anthony Solecki

Jiahuan He

Geoffrey Michael Ward

Ozan Sayin

Derek Wilcox

Esfandiar Alizadeh

Evan George Kriminger

Armen Avedis Donigian

Vishwaesh Rajiv

Randolph Paul Sinnott, Jr

Mark Lino Nielson

Rupali Jain

Liubo Li

Eran Dvir

Carlos Alberta Huertas Villegas

Systems and methods for generating and processing modeling workflows are disclosed. In some examples, a reference distribution of scores generated by a model is determined. The reference distribution of scores is recorded in a structured database. One or more unexpected scores are detected during execution of the model. To detect the one or more unexpected scores, a production distribution of scores is compared with the reference distribution of scores recorded in the structured database. The production distribution of scores is generated by the model for a production input data set. An alert is then provided to an external system, when an alert condition is determined to be satisfied based on the comparison. The alert indicates detection of the one or more unexpected scores.

Systems and methods for enriching modeling tools and infrastructure with semantics

Systems and methods for training models to improve fairness.

Systems and methods for model fairness

Methods, non-transitory computer readable media, and model evaluations systems for understanding diverse machine learning models (MLMs) are disclosed. In some examples, a feature contribution value is determined for features included in a reference or evaluation input data set. The evaluation input data set represents a protected class population and each feature contribution value identifies a contribution by a feature to a difference in output generated by an MLM for the evaluation input data set. Model explanation information is generated using the feature contribution values and execution of the MLM is monitored. The model explanation information explains the difference in output generated by the MLM for the evaluation input data set and includes information relating to a model-based decision. A report is generated from a knowledge graph for the MLM and output via a GUI to an operator device that includes the model explanation information.

Systems and methods for decomposition of differentiable and non-differentiable models

Systems and methods for generating and processing modeling workflows.

Systems and methods for understanding diverse machine learning models.

The subject invention can track and apply user edits to a source document as a sequence of changes. The changes can be applied in a document or spatial order irrespective of temporal factors. The invention can maintain intervals that represent user operations (e.g., insertions, deletions, zero-net-length changes). As well, the invention can infer a location in the original document that corresponds to a particular operation. In accordance therewith, the invention can arrange temporally sequenced user document modifications into an order consistent with the layout of the document file encoding. This functionality of mapping re-sequenced changes into the original document data representation is one novel feature of the invention. The invention can enable portions of the source document loaded into memory on an as-needed basis whereby changes relevant to the instant portion can be made.

Memory optimizing for re-ordering user edits

An animation system provides synchronization services to synchronize actions of two more interactive user interface characters that are displayed simultaneously. The animation services allow applications to make animation requests to control the actions of characters on the display. These actions include playing one of the character's animation sequences and generating speech output with lip-synched animation of the character's mouth. Accessible via script commands or an Application Programming Interface, the synchronization services allow an application to control interaction between two or more characters on the display. Applications can synchronize actions by invoking straightforward commands such as Wait, Interrupt, or Stop. In response to these commands, the animation server synchronizes scheduled actions by halting playback of a character until a specified action of another character completes or halting a specified action of one character after scheduled actions for another character are completed.

Method and system for synchronizing scripted animations

The animation of a speaking character is synchronized with recorded speech by creating and playing a linguistically enhanced sound file. A sound editing tool employs a speech recognition engine to create the linguistically enhanced sound file from recorded speech and a text of the speech. The speech recognition engine provides timing information related to word breaks and phonemes that is used by the sound editing tool to annotate the speech sound data when creating the linguistically enhanced sound file. When the linguistically enhanced sound file is played to produce sound output, the timing information is retrieved to control the animated character's mouth movement and word pacing in the character's word balloon. The sound editing tool additionally provides editing functions for manipulating the timing information. A text to speech engine can use the same programming interface as the linguistically enhanced sound file player to send notifications to the animation, providing prototyping without recorded speech. Since both use the same interface, recorded speech can be incorporated at a later time with minimal modifications.

Synchronizing the moveable mouths of animated characters with recorded speech

Growing community of inventors

Redmond, WA, United States of America

John Wickens Lamb Merrill