Abstract: A user preference hierarchy is determined from user response to images. Images may be tagged using machine learning models trained to determine values for images. Products are clustered according to product vectors. Images of products within a cluster are clustered according to composition and groups of images are selected from image clusters for soliciting feedback regarding user preference for products of a cluster. Feedback is used to train a user preference model to estimate affinity for a product vector. A user may provide feedback regarding a price point and products are weighted according to a distribution about the price point. The distribution may be asymmetrical according to direction of movement of the price point. Filters may be dynamically defined and presented to a user based on popularity and frequency of occurrence of attribute-value pairs of search results and based on feedback regarding the search results.

Abstract: A user preference hierarchy is determined from user response to images. Images may be tagged using machine learning models trained to determine values for images. Products are clustered according to product vectors. Images of products within a cluster are clustered according to composition and groups of images are selected from image clusters for soliciting feedback regarding user preference for products of a cluster. Feedback is used to train a user preference model to estimate affinity for a product vector. A user may provide feedback regarding a price point and products are weighted according to a distribution about the price point. The distribution may be asymmetrical according to direction of movement of the price point. Filters may be dynamically defined and presented to a user based on popularity and frequency of occurrence of attribute-value pairs of search results and based on feedback regarding the search results.

Abstract: An integrated approach is provided to running web applications with various compatibilities that do not require third-party extensions or sacrifice user experience. When a page of an application incompatible with a first browser is to be executed, an instance of a second browser with which the page is compatible is separately invoked and/or executed without opening a window for that second browser. Instead, the second browser opens a page within the window of the first browser. Moreover, rendering and/or contextual information for a page opened via the second browser is provided to the first browser. Such information is utilized by the first browser to provide a visual experience to the user in which the web page is being hosted by the first browser, even though it is in fact being hosted by the second browser.

Abstract: An integrated approach is provided to running web applications with various compatibilities that do not require third-party extensions or sacrifice user experience. When a page of an application incompatible with a first browser is to be executed, an instance of a second browser with which the page is compatible is separately invoked and/or executed without opening a window for that second browser. Instead, the second browser opens a page within the window of the first browser. Moreover, rendering and/or contextual information for a page opened via the second browser is provided to the first browser. Such information is utilized by the first browser to provide a visual experience to the user in which the web page is being hosted by the first browser, even though it is in fact being hosted by the second browser.

Abstract: Images are tagged with values in an image data hierarchy that is most subjective at its top level and least subjective at its bottom level, such as a hierarchy including style, type, and features for clothing. A user preference hierarchy is determined from user response to images that are tagged. Tagged images may be generated by processing them with machine learning models trained to determine values for images. Product records including images and other data are analyzed to generate attribute vectors that are encoded to generate product vectors. Products are clustered according to their product vectors. Images of products within a cluster are clustered according to composition and groups of images are selected from image clusters for soliciting feedback regarding user preference for products of a cluster. Feedback is used to train a user preference model to estimate user affinity for a product having a given product vector.

Abstract: A user preference hierarchy is determined from user response to images. Images may be tagged using machine learning models trained to determine values for images. Products are clustered according to product vectors. Images of products within a cluster are clustered according to composition and groups of images are selected from image clusters for soliciting feedback regarding user preference for products of a cluster. Feedback is used to train a user preference model to estimate affinity for a product vector. A user may provide feedback regarding a price point and products are weighted according to a distribution about the price point. The distribution may be asymmetrical according to direction of movement of the price point. Filters may be dynamically defined and presented to a user based on popularity and frequency of occurrence of attribute-value pairs of search results and based on feedback regarding the search results.

Abstract: A user preference hierarchy is determined from user response to images that are tagged. Tagged images may be generated by processing them with machine learning models trained to determine values for images. Product records including images and other data are analyzed to generate attribute vectors that are encoded to generate product vectors. Products are clustered according to their product vectors. Images of products within a cluster are clustered according to composition and groups of images are selected from image clusters for soliciting feedback regarding user preference for products of a cluster. Feedback is used to train a user preference model to estimate affinity for a product having a given product vector. A user may provide feedback regarding a price point and products are weighted according to a distribution having a highest value at the price point. The distribution may be asymmetrical according to direction of movement of the price point.

Abstract: Images are tagged with values in an image data hierarchy that is most subjective at its top level and least subjective at its bottom level, such as a hierarchy including style, type, and features for clothing. A user preference hierarchy is determined from user response to images that are tagged. Tagged images may be generated by processing them with machine learning models trained to determine values for images. Product records including images and other data are analyzed to generate attribute vectors that are encoded to generate product vectors. Products are clustered according to their product vectors. Images of products within a cluster are clustered according to composition and groups of images are selected from image clusters for soliciting feedback regarding user preference for products of a cluster. Feedback is used to train a user preference model to estimate user affinity for a product having a given product vector.

Abstract: An integrated approach is provided to running web applications with various compatibilities that do not require third-party extensions or sacrifice user experience. When a page of an application incompatible with a first browser is to be executed, an instance of a second browser with which the page is compatible is separately invoked and/or executed without opening a window for that second browser. Instead, the second browser opens a page within the window of the first browser. Moreover, rendering and/or contextual information for a page opened via the second browser is provided to the first browser. Such information is utilized by the first browser to provide a visual experience to the user in which the web page is being hosted by the first browser, even though it is in fact being hosted by the second browser.

Abstract: Web pages compatible with different web browsers are seamlessly rendered within the same tab of a single web browser window. A first browser is executed in a computing device, and has an open browser window. A request is received at the first browser to navigate to a first page. The first page is determined to be incompatible with the first browser and compatible with a second browser. Under control by the first browser, an instance of the second browser is invoked without a browser window opening for the instance of the second browser. An address of the first page is provided to the instance of the second browser. The instance of the second browser renders the first page within a tab in the open browser window of the first browser.

Abstract: Various implementations described herein are directed to an integrated circuit. The integrated circuit may include read circuitry coupled to bitlines, and the read circuitry may be activated based on a read select signal to perform a read operation on the bitlines. The integrated circuit may include write circuitry coupled to the bitlines, and the write circuitry may be activated based on a write select signal to perform a write operation on the bitlines. The integrated circuit may include bitline discharge control circuitry coupled to the bitlines and the write circuitry, and the bitline discharge control circuitry may control the bitline discharge of the bitlines during the read operation so as to restrict a false read on the bitlines by providing a discharge boundary for the bitlines during the read operation.

Abstract: A dynamic attribute search engine system can be implemented to execute faceted searches using dynamic facets. In embodiments, a dynamic attribute search engine system includes one or more product databases, a data ingestor, an index, a query service, and a faceted searcher. The dynamic attribute search engine system is programmed to dynamically determine facets for a faceted search in response to dominant categories identified for each search query.

Abstract: Various implementations described herein are directed to an integrated circuit. The integrated circuit may include read circuitry coupled to bitlines, and the read circuitry may be activated based on a read select signal to perform a read operation on the bitlines. The integrated circuit may include write circuitry coupled to the bitlines, and the write circuitry may be activated based on a write select signal to perform a write operation on the bitlines. The integrated circuit may include bitline discharge control circuitry coupled to the bitlines and the write circuitry, and the bitline discharge control circuitry may control the bitline discharge of the bitlines during the read operation so as to restrict a false read on the bitlines by providing a discharge boundary for the bitlines during the read operation.

Abstract: A circuit structure is provided. The circuit structure includes first pfet device. The circuit structure further includes a first nfet device connected to the pfet device. The circuit structure further includes a keeper nfet device that reduces stress associated with the first nfet device by keeping the first nfet device off during its functional state. The circuit structure further includes a keeper pfet device that reduces stress associated with the first pfet device by keeping the first pfet device off during its functional state.

Abstract: A dynamic attribute search engine system can be implemented to execute faceted searches using dynamic facets. In embodiments, a dynamic attribute search engine system includes one or more product databases, a data ingestor, an index, a query service, and a faceted searcher. The dynamic attribute search engine system is programmed to dynamically determine facets for a faceted search in response to dominant categories identified for each search query.

Abstract: Approaches for a memory including a cell array are provided. The memory includes a first device of the cell array which is connected to a bitline and a node and controlled by a word line, and a second device of the cell array which comprises a third device which is connected to a source line and the node and controlled by the word line and a fourth device which is connected between the word line and the node. In the memory, in response to another word line in the cell array being activated and the word line not being activated to keep the first device in an unprogrammed state, the third device isolates and floats the node such that a voltage level of a gate to source of the first device is clamped down by the fourth device to a voltage level around zero volts.

Abstract: Approaches for a memory including a cell array are provided. The memory includes a first device of the cell array which is connected to a bitline and a node and controlled by a word line, and a second device of the cell array which comprises a third device which is connected to a source line and the node and controlled by the word line and a fourth device which is connected between the word line and the node. In the memory, in response to another word line in the cell array being activated and the word line not being activated to keep the first device in an unprogrammed state, the third device isolates and floats the node such that a voltage level of a gate to source of the first device is clamped down by the fourth device to a voltage level around zero volts.

Abstract: A circuit structure is provided. The circuit structure includes first pfet device. The circuit structure further includes a first nfet device connected to the pfet device. The circuit structure further includes a keeper nfet device that reduces stress associated with the first nfet device by keeping the first nfet device off during its functional state. The circuit structure further includes a keeper pfet device that reduces stress associated with the first pfet device by keeping the first pfet device off during its functional state.

Abstract: A memory device with an age-detect-and-correct (ADAC) circuit that detects skew caused by bias temperature instability fatigue (that is, bias temperature instability stress accumulated over time), and counters skew by selectively adjusting the proportion (measured temporally) of active state operation to idle state operation. Also, a memory burn-in device using a similar ADAC circuit.