Abstract: A transducer is provided for emitting and receiving acoustic waves, a method for operating and a method for producing the same. The transducer comprises a casing that forms a cavity. The casing comprises an excitation surface and an emitting surface, which is arranged opposite to the excitation surface. A transducer element is provided at the excitation surface, and an acoustic diffuser is provided at the emitting surface of the casing, wherein a diffusing structure of the acoustic diffuser faces the cavity, and wherein the diffusing structure is provided by an array of columns.

Abstract: A background process is configured to periodically scrub a boot storage of a storage controller to ensure operational correctness of the boot storage. One or more foreground processes store a system configuration data of the storage controller in the boot storage of the storage controller. The background process and the one or more foreground processes are executed to meet predetermined performance requirements for the background process and the one or more foreground processes.

Abstract: In one aspect of multi-mode address mapping management in accordance with the present disclosure, mapping and unmapping operations may be conducted in one of multiple address mapping management modes to both improve overall system performance and maintain data integrity. In one embodiment, a first address mapping management mode such as a rigorous mode, for example, confirms completion of an unmapping of an address mapped data unit buffer before a re-mapping is permitted. Mapping and unmapping operations may be switched to a performance mode in which unmap completion confirmation is bypassed to improve performance. In one embodiment, address mapping management modes may be switched in real time as a function of monitored operating conditions. Other aspects and advantages are provided, depending upon the particular application.

Abstract: Systems and methods to allow multiplexing of uplink transmissions over configured-grant resources. If a base station allocates configured-grant uplink transmission resources to a further UE the base station transmits an indication of the overlap to relevant UEs. The UEs may apply an explicit HARQ feedback process, or other techniques to manage transmissions such as transmitting a scheduling request with the data.

Abstract: Techniques and systems are described for performing semantic text searches. A semantic text-searching solution uses a machine learning system (such as a deep learning system) to determine associations between the semantic meanings of words. These associations are not limited by the spelling, syntax, grammar, or even definition of words. Instead, the associations can be based on the context in which characters, words, and/or phrases are used in relation to one another. In response to detecting a request to locate text within an electronic document associated with a keyword, the semantic text-searching solution can return strings within the document that have matching and/or related semantic meanings or contexts, in addition to exact matches (e.g., string matches) within the document. The semantic text-searching solution can then output an indication of the matching strings.

Abstract: Examples of the present disclosure describe systems and methods of providing real-time scanning of IP addresses. In aspects, input may be received by a real-time IP scanning system. The system may generate one or more work orders based on the input. A scanner associated with the system may access a work order and attempt to communicate with one or more devices identified by the work order. If the attempted communication with a device is successful, a protocol analyzer may be used to provide a predefined payload to the device. If the response from the device matches an expected string, the device may be determined to be a safe and/or legitimate device. If the response from the device does not match an expected string, the device may be determined to be a malicious device.

Abstract: A system and method for seamless exchange and interaction of multimedia content between communication devices in a network are disclosed. The method can include the discovery and identification of devices within proximity of a sending device. The found devices can be authenticated through unique identifiers established during registration. Connection requirements can be determined based on the identifiers associated with the found devices and the sending device. In turn, the sender can establish a connection with the found devices using the connection requirements. The sending device can share or serve as a remote control to redirect and navigate the content, with a simple action or a gesture command, to the found device. The shared multimedia content, can either reside on the sender's mobile device or on a remote server within a connected network.

Abstract: A plurality of dispatch queues corresponding to a plurality of processing entities are maintained, wherein each dispatch queue includes one or more task control blocks or is empty, and wherein an ordered list of dispatch queues is maintained for each processing entity of the plurality of processing entities. A state for each of the plurality of dispatch queues is determined and the determined state is compared to a desired state for the plurality of dispatch queues. A task control block is moved from one dispatch queue to another dispatch queue, in response to the comparing of the determined state to the desired state for the plurality of dispatch queues.

Abstract: Systems and processes are provided to prevent light ends such as methane, ethylene and ethane from building up in an olefin/paraffin separation system that uses a combination of a membrane and distillation column for this separation. In one embodiment a small stripper column is provided downstream from a selective hydrogenation reactor. In the other embodiment, a surge vessel with a receiver is added to the retentate stream of the membrane unit.

Abstract: A multicore optical fiber that includes a plurality of waveguiding cores disposed in a cladding. The plurality of cores are situated adjacent to at least one other core with a core center to core center spacing being not larger than 10 times the radius of the average core, such that the greater than 10% of the light will couple from one core to the adjacent core over a propagating distance of 1 cm, along the fiber length so as to provide coupling between the adjacent cores and to enable quantum walk. The plurality waveguiding cores are disposed in the cladding in a ring distribution or at least a portion of the ring distribution.

Abstract: A multicore optical fiber that includes a plurality of waveguiding cores disposed in a cladding. The plurality of waveguiding cores include one or more first waveguiding cores that have a first propagation constant and one or more second waveguiding cores that have a second propagation constant, where the first propagation constant differs from the second propagation constant. The one or more first waveguiding cores and the one or more second waveguiding cores are disposed in the cladding in a ring distribution and at least a portion of the ring distribution is arranged based on a quasi-periodic sequence having a plurality of sequence segments. Each sequence segment is determined based on a quasi-periodic function, has an order, and corresponds to an arrangement segment of a first waveguiding cores, a second waveguiding cores, or combinations thereof. The ring distribution includes at least one arrangement segment corresponding with a third-order sequence segment or higher of the quasi-periodic sequence.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media for generating pairs of natural language queries and corresponding query-language representations. For example, the disclosed systems can generate a contextual representation of a prior-generated dialogue sequence to compare with logical-form rules. In some implementations, the logical-form rules comprise trigger conditions and corresponding logical-form actions for constructing a logical-form representation of a subsequent dialogue sequence. Based on the comparison to logical-form rules indicating satisfaction of one or more trigger conditions, the disclosed systems can perform logical-form actions to generate a logical-form representation of a subsequent dialogue sequence.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for selecting hyper-parameter sets by utilizing a modified Bayesian optimization approach based on a combination of accuracy and training efficiency metrics of a machine learning model. For example, the disclosed systems can fit accuracy regression and efficiency regression models to observed metrics associated with hyper-parameter sets of a machine learning model. The disclosed systems can also implement a trade-off acquisition function that implements an accuracy-training efficiency balance metric to explore the hyper-parameter feature space and select hyper-parameters for training the machine learning model considering a balance between accuracy and training efficiency.

Abstract: A music audio control and distribution system in a location, comprising a first wireless music audio device connected to a local network, the network operably connecting the music audio device with at least one of: (1) a music audio player or (2) a music audio output device, and with a remote music audio source operably connected to the local network. The location of the local network is one of (a) a home, (b) an office, and (c) an automobile. The system also includes a second music audio device comprising at least one of (i) a music audio player or (ii) a music audio output device. The second music audio device is also connected to the local network and is controllable by the first wireless music audio device.

Abstract: An entrainment-based filtering system for use in connection with a facemask or respirator that includes: a filter housing defining an interior volume comprising at least one airflow pathway having an airflow pathway length that is longer than the length of the interior volume and spaced within the interior volume; and a plurality of entrainment substrates. The plurality of entrainment substrates is positioned within the at least one airflow pathway and the plurality of substrates and the dimensions of the at least one airflow pathway together function to entrain particles on the surface of the chemically coated entrainment substrates and prevent at least 95% of particles traveling through the at least one airflow pathway from exiting the entrainment-based filter.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating dialogue responses based on received utterances utilizing an independent gate context-dependent additive recurrent neural network. For example, the disclosed systems can utilize a neural network model to generate a dialogue history vector based on received utterances and can use the dialogue history vector to generate a dialogue response. The independent gate context-dependent additive recurrent neural network can remove local context to reduce computation complexity and allow for gates at all time steps to be computed in parallel. The independent gate context-dependent additive recurrent neural network maintains the sequential nature of a recurrent neural network using the hidden vector output.

Abstract: Methods for natural language semantic matching performed by training and using a Markov Network model are provided. The trained Markov Network model can be used to identify answers to questions. Training may be performed using question-answer pairs that include labels indicating a correct or incorrect answer to a question. The trained Markov Network model can be used to identify answers to questions from sources stored on a database. The Markov Network model provides superior performance over other semantic matching models, in particular, where the training data set includes a different information domain type relative to the input question or the output answer of the trained Markov Network model.

Abstract: Systems and methods for parsing natural language sentences using an artificial neural network (ANN) are described. Embodiments of the described systems and methods may generate a plurality of word representation matrices for an input sentence, wherein each of the word representation matrices is based on an input matrix of word vectors, a query vector, a matrix of key vectors, and a matrix of value vectors, and wherein a number of the word representation matrices is based on a number of syntactic categories, compress each of the plurality of word representation matrices to produce a plurality of compressed word representation matrices, concatenate the plurality of compressed word representation matrices to produce an output matrix of word vectors, and identify at least one word from the input sentence corresponding to a syntactic category based on the output matrix of word vectors.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer-readable media that can determine an answer to a query based on matching probabilities for combinations of respective candidate answers. For example, the disclosed systems can utilize a gated-self attention mechanism (GSAM) to interpret inputs that include contextual information, a query, and candidate answers. The disclosed systems can also utilize a memory network in tandem with the GSAM to form a gated self-attention memory network (GSAMN) to refine outputs or predictions over multiple reasoning hops. Further, the disclosed systems can utilize transfer learning of the GSAM/GSAMN from an initial training dataset to a target training dataset.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media for techniques for identifying textual similarity and performing answer selection. A textual-similarity computing model can use a pre-trained language model to generate vector representations of a question and a candidate answer from a target corpus. The target corpus can be clustered into latent topics (or other latent groupings), and probabilities of a question or candidate answer being in each of the latent topics can be calculated and condensed (e.g., downsampled) to improve performance and focus on the most relevant topics. The condensed probabilities can be aggregated and combined with a downstream vector representation of the question (or answer) so the model can use focused topical and other categorical information as auxiliary information in a similarity computation.