Abstract: This document relates to training of machine learning models. One example method involves providing a machine learning model having a first classification layer, a second classification layer, and an encoder that feeds into the first classification layer and the second classification layer. The example method also involves obtaining first training examples having explicit labels and second training examples having inferred labels. The inferred labels are based at least on actions associated with the second training examples. The example method also involves training the machine learning model using the first training examples and the second training examples using a training objective that considers first training loss of the first classification layer for the explicit labels and second training loss of the second classification layer for the inferred labels. The method also involves outputting a trained machine learning model having the encoder and the first classification layer.

Abstract: A system and method for speed control as a function of battery capability in an electric aircraft is illustrated. The system comprises a battery pack coupled to the electric aircraft, a sensor coupled to the battery pack and configured to detect a battery pack output, and a computing device coupled to the sensor. The computing device is configured to receive the battery pack output from the sensor and adjust a maximum speed of the electric aircraft as a function of the battery pack output and a battery pack output threshold.

Abstract: A fibrous insulation product having a plurality of randomly oriented glass fibers and a binder composition that holds the glass fibers together is disclosed. The fibrous insulation product has an R-value in the range of 10 to 54 and, after curing, has a density, when uncompressed, in the range of 0.30 pcf to 2.7 pcf. Furthermore, the fibrous insulation product includes glass fibers that, prior to the application of the binder composition, have an average fiber diameter in the range of 15 HT to 19 HT and a quantity of binder that is in the range of 2% to 10% by weight of the fibrous insulation product. The fibrous insulation product also has an average fiber diameter to density ratio (Fd/D) of less than or equal to 40 and a comfort factor less than or equal to 3.417(Fd/D)+60.

Abstract: A non-invasive thermal dispersion flow meter with chronometric monitor for fluid leak detection includes a heater, an ambient temperature sensor and a flow rate sensor which are configured to sense the temperature of a fluid in a conduit, and then monitor the flow of that fluid through the conduit. Based upon the ambient temperature sensor readings, the flow rate sensor and heater may be adjusted to optimize the operation of the system to detect leaks. Based on the sensor readings, the flow may be adjusted to prevent damage and leaks by relieving the system of excess pressure. Geographic location, occupancy sensors and occupant identifiers are used to control the system to facilitate operation and minimize leak damage when occupants are away.

Abstract: Optical thin film designs are provided that achieve significantly improved laser damage thresholds and ultra-low-loss. These advances may be achieved by utilizing materials with electronic band gaps and refractive indices that are higher than those that are conventionally used.

Abstract: An osteotome having a shaft assembly that includes a guide extending in a longitudinal direction of the shaft assembly, and a blade adjacent the guide and extending in the longitudinal direction of the shaft assembly.

Abstract: A data access recovery apparatus includes: a transceiver; a memory; and a processor communicatively coupled to the transceiver and the memory and configured to: receive a request to restore backed-up unstructured data files associated with the request; send active data files, of the backed-up unstructured data files, to a data-access server in response to receiving the request; receive an indication of a particular data file of the backed-up unstructured data files; and send, in response to receiving the indication, the particular data file to the data-access server before the particular data file would be sent, if at all, absent receiving the indication.

Abstract: The technology relates to cosharding tables within a distributed storage system. A data table including one or more rows may be received. Each row in the data table may include an identifier key and pieces of data. Each piece of data in the data table may be indexed into individual rows of an index table, wherein each row in the index table includes data associated with the identifier key of the data table from which the piece of data in the respective row was indexed. The index table may be sharded into splits, wherein the sharding includes assigning each row of the index table into one of the splits based on the identifier key of the data table from which the piece of data in the respective row was indexed. The splits may be stored into two or more portions of the distributed storage system.

Abstract: The present disclosure relates to compartment system for storing and organizing various items and objects. The compartment system includes an exemplary divider which fits within pre-existing boxes and separates the same into various storage compartments. The divider is generally made of one or more sidewalls, one or more transverse divider walls, and optionally one or more axial divider walls. Methods of forming the exemplary dividers are also disclosed.

Abstract: Disclosed is an antimicrobial coating and antimicrobial laminate that can be used on surfaces to prevent the spread of bacteria and viruses. Spray-on techniques can be used with antimicrobial agents mixed with an air-drying antimicrobial coating and a heated liquid plastic antimicrobial coating. Efficacy of the antimicrobial coating can be monitored by detecting changes in texture, color, opaqueness, fluorescence, and other techniques. Soft plastics and acrylics can be used to suspend the antimicrobial agents to provide better efficacy in the release of antimicrobial agents over time.

Abstract: Presented are oleophobic surface treatments for electric machines, methods for making/using such electric machines, and vehicles employing traction motors having oleophobic treatments on select “non-target” surfaces. An electric machine includes a direct-cooling thermal management system that circulates a coolant fluid to the electric machine's outer housing. A stator assembly, which is attached to the housing, includes a stator core with one or more electromagnetic windings mounted to the stator core. A rotor assembly is rotatably mounted to the hosing adjacent the stator assembly. The rotor assembly includes a rotor core with one or more magnets mounted to the rotor core and spaced across an air gap from the winding(s). Select components of the outer housing, rotor assembly, and/or stator assembly have a target surface with an oleophobic surface treatment that reduces the non-target surface's wetted area and decreases the mass of coolant fluid contacting the non-target surface.

Abstract: The present disclosure relates generally to determining a language for speech recognition of a spoken utterance, received via an automated assistant interface, for interacting with an automated assistant. The system can enable multilingual interaction with the automated assistant, without necessitating a user explicitly designate a language to be utilized for each interaction. Selection of a speech recognition model for a particular language can based on one or more interaction characteristics exhibited during a dialog session between a user and an automated assistant. Such interaction characteristics can include anticipated user input types, anticipated user input durations, a duration for monitoring for a user response, and/or an actual duration of a provided user response.

Abstract: Genetically engineered Foot and Mouth Disease Virus (FMDV) and related engineered proteins and polynucleotides, nanolipoprotein particles, compositions, methods and systems are described. The genetically engineered FMDV is modified by the strategic insertion of a protein tag into select regions of the FMDV genome which encode viral proteins that are exposed on the surface of the FMDV viral capsid. The inserted protein tag is displayed as a decoration or attachment on the viral capsid surface.

Abstract: Methods, apparatus, and processor-readable storage media for block-level classification of unstructured data are provided herein. An example apparatus includes a host device comprising a processor coupled to memory, the host device being configured to communicate over a network with a storage system, and further being configured to: obtain a pointer to a page cache associated with an input-output operation for at least one page of unstructured data of a file; obtain an index node object of the file based at least in part on the pointer to the page cache; derive at least one characteristic of the file based at least in part on the obtained index node object; and provide an indication of the at least one characteristic to the storage system. The storage system determines whether to apply one or more functions to the unstructured data based on the indication.

Abstract: A code notebook and backend cloud service are configured to intelligently analyze program source code that a developer wants analyzed. A user drafts a code query to be answered about the source code that may specify specific variables, code structure elements, and/or program flows to be scrutinized. A cloud-computing environment builds a code database of the source code and analyzes its text, code structures, and program flows. The code database is embedded with indications of semantic equivalences for text in the source code, identifications of different code structural elements, and program flows. In the cloud-computing environment, a query service takes the code query of the developer and queries the database with machine-learned embeddings, generating query results that are shared with the developer and shown in a representation of the source code.

Abstract: Compounds of Formula (I) are provided herein. Such compounds, as well as pharmaceutically acceptable salts and compositions thereof, are useful for treating diseases or conditions, including conditions characterized by excessive cellular proliferation, such as cancer and tumors, as well as viral infections such as HIV.

Abstract: A system, method, and computer-readable medium for determining customer adoption based on monitored data, including receiving product usage parameters from a product data store on the computer network, each product usage parameter being determined based on tracking usage of the product by the customer over a predetermined time period, storing a customer profile for the customer comprising customer parameters, the customer parameters being determined based on customer information stored in a customer database on the computer network, receiving service parameters from a customer support data store on the computer network, each service parameter being determined based on tracking support services provided to the customer for the product over the predetermined time period, and generating a product adoption score by applying a machine learning model to the product usage parameters and the customer profile to generate a usage-based adoption score and adjusting the usage-based adoption score based on the service par

Abstract: Non-limiting embodiments of the present disclosure relate to a method comprising: obtaining asphalt shingle waste (ASW) and performing grinding, screening, and separating steps on the ASW. In some embodiments, granules are removed from the ASW. In some embodiments, the method transforms ASW into ASW powder. In some embodiments, the ASW powder is formed into a plurality of briquettes. In some embodiments, at least one of: the ASW powder, the plurality of briquettes, or any combination thereof is fed into a mixing process that results in an ASW powder filled coating.

Abstract: A cross-linkable powder for use in a selective laser sintering (SLS) process for additive manufacturing is disclosed as well as a novel manufacturing process to form a 3D object using said cross-linkable powder. The manufacturing process makes it possible to create interlayer covalent bondings between deposited layers of cross-linkable powder such that 3D printed objects are achieved having improved mechanical strength, less object deformation and/or no warping.

Abstract: This invention pertains to protecting communications between multiple sensors and emitters or securing data transmission between multiple computers or multiple vehicles. This invention provides a secure method for two or more parties to communicate privately, even when the processor has malicious malware or there is a backdoor in the main processor. In some embodiments, the energy received by the sensor is encrypted before it undergoes an analog to digital conversion. In some embodiments, the encryption occurs inside the sensor. In some embodiments, the encryption hardware is a part of the sensor and creates unpredictable energy changes that interact with the sensor. In some embodiments, there are less than 40 sensors in a communication system and in other embodiments there are more than 1 billion sensors. In some embodiments, the invention provides a method for the sensors of a network of self-driving cars to communicate securely.