Abstract: Described are techniques for predicting when data associated with a user input is likely to be selected for deletion. The system may use a trained model to assist with such predictions. The trained model can be configured based on deletions associated with a user profile. An example process can including receiving user input data corresponding to the user profile, and processing the user input data to determine a user command. Based on characteristic data of the user command, the trained model can be used to determine that data corresponding to the user command is likely to be selected for deletion. The trained model can be iteratively updated based on additional user commands, including previously received user commands to delete user input data.

Abstract: Systems and methods for dynamically relocating pods to optimize inter-pod networking efficiency are provided. The method comprises receiving and storing inter-pod traffic data for a plurality of pods. The plurality of pods includes a first pod, a second pod, and a third pod. The method further includes receiving and storing node resource availability data for each node of a plurality of nodes, generating a queue that sorts the plurality of pods by an amount of inter-pod traffic indicated by the inter-pod traffic data, generating a hash that maps one or more parameters to the plurality of nodes, selecting, based on the generated hash, a node of the plurality of nodes, and dynamically relocating a highest ranked pod of the plurality of pods from the generated queue to the selected node.

Abstract: A system for generating and distributing a digital mixtape. In one example, the system can receive a user command to generate a digital mixtape including a user-defined compilation of music. The user command identifies a recipient of the digital mixtape and identifies one or more media content items to be included in the music compilation for the recipient. The digital mixtape can also include audio recordings from the user to be added to the digital mixtape.

Abstract: Systems and methods for dynamically skipping a presentation of a portion of frames to catch up to live based on a user selection is disclosed herein. For example, a streaming application generates a content item for live streaming where the content item comprises a plurality of frames. In response to determining that playing of the content item lags behind the live streaming of the content item, the streaming application presents a selectable option for a user input where the option comprises (a) catch up to live in a particular time period. In response to receiving the user input via a corresponding option to (a) catch up to live in the particular time period, the streaming application skips a presentation of a portion of the plurality of frames, such that the playing of the content item catches up to live in the particular time period.

Abstract: Automated risk management techniques in an information processing system are disclosed. For example, for a given item type obtainable from two or more sources, wherein each of the two or more sources has an aging policy associated with the item type that is different with respect to one another, the method predicts a quantity of the item type obtainable from each of the two or more sources that is at risk during a given future time period based on the aging policy of each of the two or more sources. The method then determines one or more actions to be initiated to mitigate the quantity of the item type obtainable from each of the two or more sources that is at risk during the given future time period.

Abstract: A method includes identifying a sequence of transformations to be performed on an input dataset via a user interface. The method also includes identifying a first context associated with the input dataset. The method further includes selecting a first one of multiple execution engines to be used to perform the sequence of transformations on the input dataset based on the first context. In addition, the method includes providing first code implementing the sequence of transformations to the first execution engine and executing the first code using the first execution engine to perform the sequence of transformations on the input dataset.

Abstract: A system including sensors and a controller for cooperative escape zone detection for a group of vehicles are provided. The sensors obtain driving condition information indicating driving environments and vehicle conditions for the group of vehicles. For each vehicle, the controller determines, based on the driving environment of the vehicle, one or more distances associated with the vehicle that are between the vehicle and one or more obstacles that surround the vehicle and determines an escape zone status for the vehicle based on the one or more distances and the driving environment and the vehicle condition of the vehicle. When the escape zone status of one in the group of vehicles fails to satisfy a pre-defined condition, the controller sends one or more control signals to one or more vehicles in the group of vehicles to create an additional escape zone for the one in the group of vehicles.

Abstract: A method includes generating, using an audio encoder, a higher-order feature representation for each acoustic frame in a sequence of acoustic frames; generating, using a decoder, based on the higher-order feature representation, a plurality of speech recognition hypotheses, each hypotheses corresponding to a candidate transcription of an utterance and having an associated first likelihood score; generating, using an external language model, for each speech recognition hypothesis, a second likelihood score; determining, using a learnable fusion module, for each speech recognition hypothesis, a set of fusion weights based on the higher-order feature representation and the speech recognition hypothesis; and generating, using the learnable fusion module, for each speech recognition hypothesis, a third likelihood score based on the first likelihood score, the second likelihood score, and the set of fusion weights, the audio encoder and decoder trained using minimum additive error rate training in the presence of t

Abstract: A utility-scale energy storage and conversion system can operate two or more inverter groups such that their reactive power commands are proportional to their available reactive power range. The control system can therefore distribute the reactive power commands in proportion to the available Q range, thereby ensuring that all inverters in the utility-scale energy storage and conversion system 100 operate with the same Q “headroom”. In addition, the utility-scale energy storage and conversion system can use an on-load tap changer (LTC) to adjust a terminal voltage associated with a first group of inverters and a second group of inverters. The first group of inverters can be associated with a first rating and the second group of inverters can be associated with a second rating that is greater than the first rating.

Abstract: One example includes a circuit. The circuit includes a first transistor having a first control terminal, a first current terminal, and a second current terminal. The first control terminal can be a first input to the circuit. The circuit also includes a second transistor having a second control terminal, a first current terminal, and a second current terminal. The second control terminal can be a second input to the circuit. The circuit also includes an adaptive bias current source coupled to the second current terminal of the respective first and second transistors. The circuit further includes a voltage offset generator coupled in parallel with the second transistor.

Abstract: A method includes obtaining a first data model from a data exploration phase performed in a first environment, where the first data model includes first metadata. The method also includes obtaining a second data model from the data exploration phase performed in a second environment different from the first environment, where the second data model includes second metadata. The method further includes generating a third data model including one or more software artifacts using the first metadata and the second metadata. Each of the one or more software artifacts is configured as one or more files that are configured for execution of at least one artificial intelligence (AI)/machine learning (ML) application.

Abstract: An autonomous system and method of comprehensively testing serverless applications. The system is configured to automatically generate test scripts and test data based on deployed and modified function code and configuration files. The system is configured to work in multi-cloud environments and is able to accommodate any cloud provider. Implementation of such a system can eliminate manual errors. Furthermore, the proposed embodiments are configured to integrate with already existing devOps pipelines for rapid test execution, and can continue automated testing in real-time as modifications to the app are made.

Abstract: A method, system, and computer program product generate, with a payment network, a first value (a) and a second value (ga), the second value (ga) generated based on the first value (a) and a generator value (g); generate, with the payment network, a plurality of random merchant numbers (mi) for a respective plurality of merchant banks; determine, with the payment network, a merchant product (M) based on a product of the plurality of random merchant numbers (mi); generate, with the payment network, a public key (pki) based on the second value (ga), the merchant product (M), and the random merchant number (mi) and a random key (rki) based on the merchant product (M) and the random merchant number (mi) for each respective merchant bank; and communicate, with the payment network, the public key (pki) and the random key (rki) to at least one respective merchant bank.

Abstract: An adaptive prefetcher for a shared system cache of a processing system including multiple requestors having a cache miss monitor and a prefetch controller. The cache miss monitor monitors requests for information from memory and identifies one of the requestors for which an identified cache line is requested. The prefetch controller submits an adaptive request for a subsequent cache line. The subsequent cache line may be determined based on a latency comparison between a loop latency (LL) of the prefetch controller and a stream latency (SL) of the identified requestor. A latency memory may be included that stores stream latencies for the requestors. The latency comparison may be used to determine how many cache lines to skip relative to the identified cache line, such as according to SL*SK<LL?SL*(SK+1) in which SK is the number of cache lines to skip.

Abstract: A method includes identifying one or more transformations to be applied in order to generate one or more features or feature sets. The method also includes generating metadata identifying the one or more features or feature sets and the one or more transformations. The method further includes using the metadata to determine the one or more features or feature sets for specified data and storing the one or more determined features or feature sets in a feature store. In addition, the method includes outputting at least some of the one or more determined features or feature sets or data associated with the at least some of the one or more determined features or feature sets from the feature store to at least one machine learning model.

Abstract: The present invention relates to nucleic acid constructs comprising methyl CpG binding protein 2 (MeCP2) promoter sequences. The present invention further relates to vectors, viral vector host cells and pharmaceutical compositions comprising said nucleic acid constructs. The present invention also concerns the therapeutic use of said nucleic acid constructs, vectors, viral vectors and pharmaceutical compositions.

Abstract: A method of text-only and semi-supervised training for deliberation includes receiving training data including unspoken textual utterances that are each not paired with any corresponding spoken utterance of non-synthetic speech, and training a deliberation model that includes a text encoder and a deliberation decoder on the unspoken textual utterances. The method also includes receiving, at the trained deliberation model, first-pass hypotheses and non-causal acoustic embeddings. The first-pass hypotheses is generated by a recurrent neural network-transducer (RNN-T) decoder for the non-causal acoustic embeddings encoded by a non-causal encoder. The method also includes encoding, using the text encoder, the first-pass hypotheses generated by the RNN-T decoder, and generating, using the deliberation decoder attending to both the first-pass hypotheses and the non-causal acoustic embeddings, second-pass hypotheses.

Abstract: An electronic gaming machine includes a display, a player input device, a credit input mechanism, and a processor programmed to perform operations comprising: (i) display a search pattern matrix to a player on the display, the search pattern matrix includes a plurality of cells that are selectable by the player; (ii) receive an input selection of cells of the search pattern matrix from the player; (iii) search a pay table associated with a wagering game provided by the gaming machine for winning patterns included within the pay table and having at least the same plurality of cells daubed as are indicated by the search pattern matrix; and (iv) display the winning patterns identified by the searching to the player via the display, the displaying including presenting the winning patterns in a matrix display format identifying one or more daubed cells of the associated winning patterns within the matrix display

Abstract: Described herein is a sensor probe for association with a portion of an aircraft. The sensor probe includes a microphone assembly having a portion configured to receive audio signals. The sensor probe further includes a nosecone associated with the microphone assembly. The nosecone assembly is configured to shield the portion of the microphone assembly from noise generated by direct impact of an airflow for a plurality of local flow angles.

Abstract: Systems, apparatus, and methods are described for delivering a therapeutic substance to a target area within or proximate to a nasal cavity of a subject, including a reservoir configured to contain the therapeutic substance and a delivery interface by which the therapeutic substance is delivered to the target area. In some embodiments, systems, apparatus, and methods described herein can deliver a therapeutic substance using iontophoresis and/or electroosmosis.