Abstract: A photovoltaic array fault diagnosis method based on composite information is provided. The method includes: collecting and preprocessing composite information data of photovoltaic array working state, including image data and text data; using the image data of photovoltaic array working state to train a pre-established fault classification model of deep convolutional neural network, to thereby obtain an image fault classification model; using the text data of photovoltaic array working state to train a pre-established fault classification model based on a support vector machine, to thereby obtain a text fault classification model; fusing the image fault classification model and the text fault classification model by logistic regression algorithm to obtain a fusion model, and training the fusion model using the composite information data of photovoltaic array working state to thereby obtain the photovoltaic array fault diagnosis model.

Abstract: Package devices and methods of manufacture are discussed. In an embodiment, a method of manufacturing an integrated circuit device includes: forming an optical device layer; forming an optical layer on the optical device layer; after the forming the optical layer, forming a first opening in the optical layer; and embedding a reflective structure in the first opening.

Abstract: Embodiments of the present disclosure relate to methods, devices, and computer readable medium for hybrid automatic repeat request (HARQ). A method in a terminal device comprises: receiving a scheduling message from a network device, the scheduling message including information on HARQ feedback, wherein the information on HARQ feedback indicates at least one of: a processing identification for HARQ feedback and the accumulated number of scheduling messages that have been transmitted for the terminal device or the processing identification; detecting data from the network device according to the scheduling message; and transmitting HARQ feedback to the network device based on the information on HARQ feedback.

Abstract: A method, system, and computer program product for reconstructing training data and building a new incremental learning model with the reconstructed training data that can be further trained. The method may include receiving new data to be inputted into a previously trained machine learning model, where the previously trained machine learning model has inaccessible training data. The method may also include generating simulated training data using a reverse form of the previously trained machine learning model. The method may also include verifying the simulated training data. The method may also include creating a new machine learning model using the simulated training data, where the new machine learning model includes a same structure as the previously trained machine learning model. The method may also include inputting the new data into the new machine learning model, where the new machine learning model is further trained with the new data.

Abstract: A device comprise a first semiconductor channel layer over a substrate, a second semiconductor channel layer over the first semiconductor channel layer, and source/drain epitaxial structures on opposite sides of the first semiconductor channel layer and opposite sides of the second semiconductor channel layer. A compressive strain in the second semiconductor channel layer is greater than a compressive strain in the first semiconductor channel layer. The source/drain epitaxial structures each comprise a first region interfacing the first semiconductor channel layer and a second region interfacing the second semiconductor channel layer, and the first region has a composition different from a composition of the second region.

Abstract: A semiconductor device includes a first layer that includes a first semiconductor material disposed on a semiconductor substrate, and a second layer of a second semiconductor material disposed on the first layer. The semiconductor substrate includes Si. The first semiconductor material and the second semiconductor material are different. The second semiconductor material is formed of an alloy including a first element and Sn. A surface region of an end portion of the second layer at both ends of the second layer has a higher concentration of Sn than an internal region of the end portion of the second layer. The surface region surrounds the internal region.

Abstract: A semiconductor device includes a first layer that includes a first semiconductor material disposed on a semiconductor substrate, and a second layer of a second semiconductor material disposed on the first layer. The semiconductor substrate includes Si. The first semiconductor material and the second semiconductor material are different. The second semiconductor material is formed of an alloy including a first element and Sn. A surface region of an end portion of the second layer at both ends of the second layer has a higher concentration of Sn than an internal region of the end portion of the second layer. The surface region surrounds the internal region.

Abstract: A device comprises a plurality of nanosheets, source/drain stressors, and a gate structure wrapping around the nanosheets. The nanosheets extend in a first direction above a semiconductor substrate and are arranged in a second direction substantially perpendicular to the first direction. The source/drain stressors are on either side of the nanosheets. Each of the source/drain stressors comprises a first epitaxial layer and a second epitaxial layer over the first epitaxial layer. The first and second epitaxial layers are made of a Group IV element and a Group V element. An atomic ratio of the Group V element to the Group IV element in the second epitaxial layer is greater than an atomic ratio of the Group V element to the Group IV element in the first epitaxial layer.

Abstract: Techniques are provided for mapping natural language to code segments. In one embodiment, the techniques involve receiving a document and software code, wherein the document comprises a natural language description of a use of the code, generating, via a vectorization process performed on the document, at least one vector or word embedding, generating, via a natural language processing technique performed on the at least one vector or word embedding, a first label set, generating, via a machine learning analysis of the software code, a second label set, determining, based on a comparison of the first label set and the second label set, a match confidence between the document and the software code, wherein the match confidence indicates a measure of similarity between the first label set and the second label set, and upon determining that the match confidence exceeds a predefined threshold, mapping the document to the software code.

Abstract: The present disclosure is directed to semiconductor structures with source/drain epitaxial stacks having a low-melting point top layer and a high-melting point bottom layer. For example, a semiconductor structure includes a gate structure disposed on a fin and a recess formed in a portion of the fin not covered by the gate structure. Further, the semiconductor structure includes a source/drain epitaxial stack disposed in the recess, where the source/drain epitaxial stack has bottom layer and a top layer with a higher activated dopant concentration than the bottom layer.

Abstract: A device includes plural semiconductor fins, a gate structure, an interlayer dielectric (ILD) layer, and an isolation dielectric. The gate structure is across the semiconductor fins. The ILD surrounds the gate structure. The isolation dielectric is at least between the semiconductor fins and has a thermal conductivity greater than a thermal conductivity of the ILD layer.

Abstract: Techniques are provided for mapping natural language to code segments. In one embodiment, the techniques involve receiving a document and software code, wherein the document comprises a natural language description of a use of the code, generating, via a vectorization process performed on the document, at least one vector or word embedding, generating, via a natural language processing technique performed on the at least one vector or word embedding, a first label set, generating, via a machine learning analysis of the software code, a second label set, determining, based on a comparison of the first label set and the second label set, a match confidence between the document and the software code, wherein the match confidence indicates a measure of similarity between the first label set and the second label set, and upon determining that the match confidence exceeds a predefined threshold, mapping the document to the software code.

Abstract: The present disclosure is directed to semiconductor structures with source/drain epitaxial stacks having a low-melting point top layer and a high-melting point bottom layer. For example, a semiconductor structure includes a gate structure disposed on a fin and a recess formed in a portion of the fin not covered by the gate structure. Further, the semiconductor structure includes a source/drain epitaxial stack disposed in the recess, where the source/drain epitaxial stack has bottom layer and a top layer with a higher activated dopant concentration than the bottom layer.

Abstract: The present invention provides a compound of Formula (I) or a pharmaceutically acceptable salt thereof; a method for manufacturing the compounds of the invention, and its therapeutic uses. The present invention further provides a combination of pharmacologically active agents and a pharmaceutical composition.

Abstract: A method, system, and computer program product for implementing automated cognitive software application error detection is provided. The method includes receiving data associated with model based self-learning software code. The annotated data is automatically divided with respect to specified categorization and grouping attributes and categorized groups comprising portions of the annotated data are generated and analyzed. At least one incorrect annotation associated a group of the categorized groups is detected and filtered. Likewise, a correct annotation for the group is detected and retrieved from a database. The correct annotation is appended to the group.

Abstract: The present invention provides a compound of Formula (I) or a pharmaceutically acceptable salt thereof; (I) a method for manufacturing the compounds of the invention, and its therapeutic uses. The present invention further provides a combination of pharmacologically active agents and a pharmaceutical composition.

Abstract: A method and device for removing an electromagnetic core, the method including: using an electromagnet to magnetize or demagnetize a metallic upper nozzle when a magneto-conductive workpiece is cut off in a WEDM manner; attracting a core capable of being completely cut off and separated in the workpiece; utilizing the metallic upper nozzle to detect whether attracted; if the core is attracted, moving the core to a target area; demagnetizing and dropping the core in a trash area. The device is applied to a WEDM machine; after the metallic magneto-conductive upper nozzle is magnetized by the electromagnet, the upper nozzle is used to attract the magneto-conductive core; a metallic water spray cover is utilized to detect whether the core is attracted; the core is moved to the target area by a motion system of the WEDM machine, and dropped in a trash device after the upper nozzle is demagnetized.

Abstract: A method includes forming a first semiconductor layer over a substrate; forming a second semiconductor layer over the first semiconductor layer; forming a dummy gate structure over the second semiconductor layer; performing an etching process to form a recess in the first and second semiconductor layers; forming a epitaxy structure over in the recess, wherein the epitaxy structure is in contact with the first and second semiconductor layers; performing a solid phase diffusion process to form a doped region in the epitaxy structure, in which the doped region is in contact with the second semiconductor layer and is separated from the first semiconductor layer; and replacing the dummy gate structure with a metal gate structure.

Abstract: A device comprises a plurality of nanosheets, source/drain stressors, and a gate structure wrapping around the nanosheets. The nanosheets extend in a first direction above a semiconductor substrate and are arranged in a second direction substantially perpendicular to the first direction. The source/drain stressors are on either side of the nanosheets. Each of the source/drain stressors comprises a first epitaxial layer and a second epitaxial layer over the first epitaxial layer. The first and second epitaxial layers are made of a Group IV element and a Group V element. An atomic ratio of the Group V element to the Group IV element in the second epitaxial layer is greater than an atomic ratio of the Group V element to the Group IV element in the first epitaxial layer.

Abstract: Filtering incoming calls according to predicted preferences of a user. User preferences are predicted by analysis of user behavior, online activity, oral queues, and purchasing history. Data analysis includes weighting caller and user attributes according to a scheme that is dynamically updated by applying user feedback and/or machine learning processes.