Microsoft Technology Licensing, LLC

Keng-hao Chang

Ruofei Zhang

Changbo Hu

Zi Yin

Qun Li

John Weston Hughes

Tianchuan Du

Paul Liu

Jia Le He

Srinivasa Reddy Neerudu

Shuangfei Zhai

Xiaozong Wang

Sujith Vishwajith

Neural network-based categorization can be improved by incorporating graph neural networks that operate on a graph representing the taxonomy of the categories into which a given input is to be categorized by the neural network based-categorization. The output of a graph neural network, operating on a graph representing the taxonomy of categories, can be combined with the output of a neural network operating upon the input to be categorized, such as through an interaction of multidimensional output data, such as a dot product of output vectors. In such a manner, information conveying the explicit relationships between categories, as defined by the taxonomy, can be incorporated into the categorization. To recapture information, incorporate new information, or reemphasize information a second neural network can also operate upon the input to be categorized, with the output of such a second neural network being merged with the output of the interaction.

Neural network categorization accuracy with categorical graph neural networks

Described herein are technologies related to constructing supplemental content items that summarize electronic landing pages. A sequence to sequence model that is configured to construct supplemental content items is trained based upon a corpus of electronic landing pages and supplemental content items that have been constructed by domain experts, wherein each landing page has a respective supplemental content item assigned thereto. The sequence to sequence model is additionally trained using self critical sequence training, where estimated click through rates of supplemental content items generated by the sequence to sequence model are employed to train the sequence to sequence model.

Generating electronic summary documents for landing pages

The present application describes a system and method for converting a natural language query to a standard query using a sequence-to-sequence neural network. As described herein, when a natural language query is receive, the natural language query is converted to a standard query using a sequence-to-sequence model. In some cases, the sequence-to-sequence model is associated with an attention layer. A search using the standard query is performed and various documents may be returned. The documents that result from the search are scored based, at least in part, on a determined conditional entropy of the document. The conditional entropy is determined using the natural language query and the document.

Query rewriting and interactive inquiry framework

A computer-implemented technique uses one or more neural networks to identify at least one item name associated with an input image using a multi-modal fusion approach. The technique is said to be multi-modal because it collects and processes different kinds of evidence regarding each detected item name. The technique is said to adopt a fusion approach because it fuses the multi-modal evidence into an output conclusion that identifies at least one item name associated with the input image. In one example, a first mode collects evidence by identifying and analyzing regions in the input image that are likely to include item name-related information. A second mode collects and analyzes any text that appears as part of input image itself. A third mode collects and analyzes text that is not included in the input image itself, but is nonetheless associated with the input image.

Processing image-bearing electronic documents using a multimodal fusion framework

A computer-implemented technique uses a generative adversarial network (GAN) to jointly train a generator neural network ('generator') and a discriminator neural network ('discriminator'). Unlike traditional GAN designs, the discriminator performs the dual role of: (a) determining one or more attribute values associated with an object depicted in input image fed to the discriminator; and (b) determining whether the input image fed to the discriminator is real or synthesized by the generator. Also unlike traditional GAN designs, an image classifier can make use of a model produced by the GAN's discriminator. The generator receives generator input information that includes a conditional input image and one or more conditional values that express desired characteristics of the generator output image. The discriminator receives the conditional input image in conjunction with a discriminator input image, which corresponds to either the generator output image or a real image.

Multi-task GAN, and image translator and image classifier trained thereby

A taxonomy of categories, attributes, and values can be conflated with new data triplets by identifying one or more conflation candidates among the attribute-value pairs within a category of the taxonomy that matches the category of the data triplet, and determining a suitable merge action for conflating the data triplet with each conflation candidate. The task of determining merge actions may be cast as a classification problem, and may be solved by an ensemble classifier.

Taxonomy enrichment using ensemble classifiers

Technologies are described herein that relate to identifying supplemental content items that are related to objects captured in images of webpages. A computing system receives an indication that a client computing device has a webpage displayed thereon that includes an image. The image is provided to a first DNN that is configured to identify a portion of the image that includes an object of a type from amongst a plurality of predefined types. Once the portion of the image is identified, the portion of the image is provided to a plurality of DNNs, with each of the DNNs configured to output a word or phrase that represents a value of a respective attribute of the object. A sequence of words or phrases output by the plurality of DNNs is provided to a search computing system, which identifies a supplemental content item based upon the sequence of words or phrases.

Pipeline for identifying supplemental content items that are related to objects in images

Images are encoded into multidimensional vectors in a high-dimensional space utilizing an image model and textual content utilizing a text model. At least one of the image model and/or the text model are tuned such that the point within the multidimensional space pointed to by a vector encoded from an image is proximate to, as determined within the context of that multidimensional space, a point pointed to by a vector encoded from correlated textual content. Received images and textual content are encoded into image vectors and text vectors, respectively, and stored in an image graph and text graph, respectively. An input image can then be encoded as an input image vector and utilized to find close vectors in both the image graph and the text graph, thereby enabling an input image to be utilized to search textual content without using classifiers to deduce textual content therefrom.

Increasing inclusiveness of search result generation through tuned mapping of text and images into the same high-dimensional space

A technique of scoring a query against a document using sequence to sequence neural networks. The technique comprises: receiving a query comprising a plurality of words from a user; performing a search for a document comprising words based on the query; feeding the words of the document as the input of an encoder of a multilayer sequence to sequence converter; generating a plurality of vectors at a decoder of the multilayer sequence to sequence converter, each vector comprising a probability associated with a respective word in the query; looking up in the respective vector each word's probability of being associated with the document; multiplying every word's probability together to determine an overall probability of the query being associated with the document; and returning the document to the user if the overall probability of the query being associated with the document is greater than a threshold value.

Dynamic tensor attention for information retrieval scoring

A computer-implemented technique is described herein for identifying one or more content items that are relevant to an input linguistic item (e.g., an input query) using a deep-structured neural network, trained based on a corpus of click-through data. The input linguistic item has a collection of input tokens. The deep-structured neural network includes a first part that produces word embeddings associated with the respective input tokens, a second part that generates state vectors that capture context information associated with the input tokens, and a third part which distinguishes important parts of the input linguistic item from less important parts. The second part of the deep-structured neural network can be implemented as a recurrent neural network, such as a bi-directional neural network. The third part of the deep-structured neural network can generate a concept vector by forming a weighted sum of the state vectors.

Growing community of inventors

San Jose, CA, United States of America

Keng-hao Chang