Abstract: Methods and systems are described herein for efficiently labeling user utterances, which may encompass any communication received from a user within a conversational interaction, and identifying novel user intents for large amounts of data. A machine learning model may be used, which is trained on embeddings of utterance data, and which may employ methods like prototypical networks and hierarchical local binary classification for hierarchical multi-label multi-class classification.

Abstract: This document generally describes systems and methods for dynamically adapting speech recognition for individual voice queries of a user using class-based language models. The method may include receiving a voice query from a user that includes audio data corresponding to an utterance of the user, and context data associated with the user. One or more class models are then generated that collectively identify a first set of terms determined based on the context data, and a respective class to which the respective term is assigned for each respective term in the first set of terms. A language model that includes a residual unigram may then be accessed and processed for each respective class to insert a respective class symbol at each instance of the residual unigram that occurs within the language model. A transcription of the utterance of the user is then generated using the modified language model.

Abstract: In some embodiments, holographically displaying items based on item compatibility with user account features may be facilitated. In some embodiments, first and second mobile devices may be detected as being within a proximity threshold of a holographic display device, where the first mobile device is associated with a first user account and the second mobile device is associated with a second user account. Based on the detection, (i) navigation information of the first and second mobile devices, (ii) return value information indicating respective return values associated with candidate items, and (iii) account feature compatibility information may be obtained and inputted into a machine learning model. Despite a first candidate item being incompatible with a second account feature set for the second mobile device, a first holographic representation of the first candidate item may be displayed.

Abstract: Methods and apparatuses for manipulating the temperature of a surface are provided. Devices of the present disclosure may include a thermal adjustment apparatus, such as a controller in electrical communication with one or more thermoelectric materials, placed adjacent to the surface of skin. The device may generate a series of thermal pulses at the surface, for providing an enhanced thermal sensation for a user. The thermal pulses may be characterized by temperature reversibility, where each pulse includes an initial temperature adjustment, followed by a return temperature adjustment, over a short period of time (e.g., less than 120 seconds). The average rate of temperature change upon initiation and upon return may be between about 0.1° C./sec and about 10.0° C./sec. In some cases, the average rate of the initial temperature adjustment is greater in magnitude than the average rate of the return temperature adjustment.

Abstract: A walking aid that assists a user in walking in an upright position. The walking aid may include a rear lower frame structure, a front lower frame structure, a first front wheel coupled to a first front lower frame member, a second front wheel coupled to a second front lower frame member, a first rear wheel coupled to the first rear lower frame member, a second rear wheel coupled to the second rear lower frame member, an upper frame structure comprising a first upper frame member, a second upper frame member, a first forearm support, and a second forearm support, a first pivot and latch mechanism pivotally coupling the first upper frame member and the first rear lower frame member together, and a second pivot and latch mechanism pivotally coupling the second upper frame member and the second rear lower frame member together.

Abstract: A curable polyorganosiloxane composition includes (a) a polyorganosiloxane having two or more curable functional group capable of causing addition reaction in one molecule; (b) a crosslinking agent having three or more crosslinking groups which have reactivity with the curable functional group of the (a) in one molecule; (c) a curing catalyst capable of catalyzing a crosslinking reaction of the (a) and (b); and (d) a silicon compound having a monovalent alkyl group of 8 to 30 carbon atoms and a hydrolyzable group which are directly bonded to a silicon atom; and in some cases, (e) scaly inorganic fine particles. An adhesive may use the curable polyorganosiloxane composition.

Abstract: A protective carrying case for an object is disclosed, wherein the protective carrying case includes a top component, a bottom component, and semi-hexagonal ends, wherein the top component and the bottom component are constructed from carbon fiber. The protective carrying case additionally includes a retaining element with vacuum split functionality to retain elements within the case.

Abstract: The invention is directed to immunogenic compositions and method of treatment comprising a peptide or nucleic acid that encodes the peptide that induces an immune response in a mammal that is protective against infection by one or more pathogens. The peptide sequence contains multiple epitopes, wherein at least one epitope is a composite epitope which is a combination of two or more conserved epitopes of the pathogen wherein the amino acid sequence of the composite is not an amino acid sequence of the pathogen. In addition, the invention is directed to vaccines comprising the peptide or nucleic acid that encodes the peptide for treating and preventing an infection in mammals such as animals and humans.

Abstract: Embodiments of the invention include an optical fiber cable. The optical fiber cable includes a multi-fiber unit tube that is substantially circular and dimensioned to receive a plurality of optical fibers. The optical fiber cable also includes a plurality of partially bonded optical fiber ribbon units positioned within the multi-fiber tube. The partially bonded optical fiber ribbon units are partially bonded in such a way that each partially bonded optical fiber ribbon is formed in a substantially circular shape or a random shape. The optical fiber cable also includes at least one elastomeric strength layer formed around the partially bonded optical fiber ribbon units. The optical fiber cable also includes an outer jacket surrounding the multi-fiber tube.

Abstract: A container lid including: a lid body which includes a flared perimeter wall, an inner structure, and a perimeter well. The inner structure defines a central throat. The perimeter well defines a plurality of openings disposed around the central throat. A lid body seal is disposed about the lid body. A lid upper includes: a lid base and a lid roof secured to the lid base. The lid base is lockably received in the central throat. The lid roof at least partially covers the perimeter well when the lid base is lockably secured in the central throat.

Abstract: Methods and devices for manipulating the temperature of a surface are generally provided. The present disclosure relates to a device including one or more heating and/or cooling elements, or other suitable thermal adjustment apparatus(es), placed near a surface, such as the skin of a user. The device may be configured to generate one or more (optionally alternating) thermal profiles at the surface, which may include a series of thermal pulses and/or essentially continuous or semi-continuous thermal input, which may vary over time. Such thermal profiles, when suitably applied, may provide enhanced thermal sensations for a user which, in some cases, may provide the user with a more pleasurable thermal experience than would otherwise be the case without the generation of the thermal profiles. In some embodiments, an alternating thermal profile may include an average frequency, an oscillation window, and/or an average temperature, each of which may be adjustable.

Abstract: According to one or more embodiments of the present disclosure, a catalyst system useful for dehydrogenation includes from 98 vol. % to 99.95 vol. % of a catalyst and from 0.05 vol. % to 2 vol. % of a combustion additive. The catalyst may include from 1 ppmw to 150 ppmw platinum, gallium, and a support material. The combustion additive may include from 150 ppmw to 1,000 ppmw platinum, gallium, and a support material. The combustion additive may include at least 1.1 times greater platinum than the catalyst.

Abstract: In some embodiments, holographic presentations despite using holographic displays having limited viewing angles may be facilitated. In some embodiments, a user may be detected via one or more sensors associated with a holographic display system, where the holographic display system being configured with a first viewing angle. Based on the detection of the user, a direction of the user may be determined. The first viewing angle of the holographic display system may also be determined. For example, the holographic display system may determine a current first viewing angle. Based on the direction of the user, the first viewing angle may be adjusted such that the user becomes within the adjusted first viewing angle of the holographic display system.

Abstract: In some aspects, a computing system may create different node embeddings (e.g., different instances of the same graph) and aggregate the node embeddings to form “multipurpose” node embeddings. As an example, the different node embeddings may include node embeddings generated using unsupervised machine learning and node embeddings generated using supervised machine learning. In this way, for example, a variety of machine learning models may use the aggregated node embeddings without the need for each machine learning model to generate separate node embeddings each time a machine learning task is performed. A machine learning model may use all or a portion of the features in the aggregated node embeddings as appropriate for the task the model is performing.

Abstract: A method to build a lookup decoder for mapping error syndromes based on quantum-stabilizer code to corresponding error corrections comprises (A) enumerating a subset of error syndromes up to a maximum error weight based on the quantum-stabilizer code; (B) iterating through the subset of error syndromes to compute an error state of highest probability for each error syndrome of the subset, where the error state defines error in a qubit register of a quantum computer; and (C) for each error syndrome of the subset of error syndromes, storing in classical computer memory an error correction based on the error state of highest probability and mapped to that error syndrome.

Abstract: A transperineal adaptor for use with a transperineal biopsy system including an ultrasound probe and biopsy guide. The transperineal adaptor includes a shaft body and a mounting body. The shaft body is configured to releasably couple to the accessory coupling portion of the probe and includes a first length, a first curvate outer surface, and an inner surface. The inner surface is configured to facilitate snapping together of the transperineal adaptor and the ultrasound probe. The mounting body extends from the shaft body and includes a second length, a second curvate surface, a first connector portion at a first end of the second curvate outer surface, and a second connector portion. The first connector portion is configured to couple to the first end of the fastener of the biopsy guide and the second connector portion configured to couple to the second end of the fastener of the biopsy guide.

Abstract: Computer-implemented methods and systems include downstream execution for individual rule-based flagging of upstream data quality errors by receiving upstream data from a plurality of sources, identifying a downstream task to be executed, applying a plurality of rules to the upstream data, generating a plurality of outputs including at least one output for each of the plurality of rules applied to the upstream data, each of the plurality of outputs being associated with a corresponding rule of the plurality of rules, identifying a tagged population based on the plurality of outputs, determining that at least one of the plurality of outputs does not meet a corresponding rule threshold, and activating the downstream execution for the tagged population after at least one of (i) updating the corresponding rule threshold or (ii) overriding an error.

Abstract: Described implementations include a contaminant-free plasma singulation process, in which residues of materials used during plasma singulation are fully removed from sidewalls of a resulting semiconductor die, without damaging the semiconductor die. From such a contaminant-free plasma singulation process, a semiconductor die may be manufactured. The semiconductor die may include a first plurality of sidewall recesses formed in a sidewall of a substrate of the semiconductor die between a first surface and a second surface of the substrate, each having at most a first depth, as well as a second plurality of sidewall recesses formed in the sidewall of the substrate and disposed between the first plurality of sidewall recesses and the second surface, each having at least a second depth that is greater than the first depth.

Abstract: This document relates to analyzing of network stack functionality that is implemented in hardware, such as on a network adapter. The disclosed implementations employ a programmable network device, such as a switch, to inject events into traffic and mirror the traffic for subsequent analysis. The events can have user-specified event parameters to test different types of network stack behavior, such as how the network adapters respond to corrupted packets, dropped packets, or explicit congestion notifications.

Abstract: Disclosed herein are system, method, and computer program product embodiments for automatically failing over all services operating on a primary region to a secondary region upon detection or notification of a fault in the primary region. When a fault exists on the primary region, the method traverses each cluster containing services operating on the primary region and prepares an input file including a list of service names identifying each service operating on the primary region. Referencing the input file, the method fails over each service from the primary region to the secondary region by modifying a service weight corresponding to each service. This failover process of services may be done simultaneously with failing over any databases corresponding to the failed-over services from the primary region to the secondary region. The method may also introduce a sleep time after modifying each service weight to avoid any potential throttling issues.