Abstract: Modules and control units cooperate to simultaneously and independently control and adjust pixel parameters non-uniformly at regional increments across an entire image captured in an image frame by pixels in a pixel array. Pixel parameter changes for pixels in a given region occur, based on i) a contextual understanding of what contextually was happening in the one or more prior image frames and ii) whether salient items are located within that region. Additionally, guidance is sent to the sensor control unit to i) increase or decrease pixel parameters within those regions with salient items and then either to i) maintain, ii) increase or iii) decrease pixel parameters within regions without salient items in order to stay within any i) bandwidth limitations ii) memory storage, and/or iii) power consumptions limitations imposed by 1) one or more image sensors or 2) the communication loop between the sensor control unit and the image processing unit.

Abstract: In general, the disclosure describes techniques for automatically generating summaries of meetings. A computing system obtains a transcript of a meeting and may produce, based on the transcript of the meeting, a data structure that comprises utterance features. Furthermore, the computing system may determine, based on the transcript of the meeting, temporal bounds of a plurality of activity episodes within the meeting. For each respective activity episode of a plurality of activity episodes, the computing system may determine, based on the utterance features associated with the respective activity episode, a conversational activity type associated with the respective activity episode. Additionally, the computing system may produce an episode summary for the respective activity episode that is dependent on the determined conversational activity type associated with the respective activity episode.

Abstract: A method for providing a real time, three-dimensional (3D) navigational map for platforms includes integrating at least two sources of multi-modal and multi-dimensional platform sensor information to produce a more accurate 3D navigational map. The method receives both a 3D point cloud from a first sensor on a platform with a first modality and a 2D image from a second sensor on the platform with a second modality different from the first modality, generates a semantic label and a semantic label uncertainty associated with a first space point in the 3D point cloud, generates a semantic label and a semantic label uncertainty associated with a second space point in the 2D image, and fuses the first space semantic label and the first space semantic uncertainty with the second space semantic label and the second space semantic label uncertainty to create fused 3D spatial information to enhance the 3D navigational map.

Abstract: A hyperdexterous surgical system is provided. The system on include one or more surgical arms coupleable in a fixture and configured to support one or more surgical tools. The system can include an electronic control system configured to communicate electronically with the one or more robotic surgical tools. The control system can electronically control the operation of the one or more surgical tools. The system can include one or more portable handheld controllers actuatable by a surgeon to communicate one or more control signals to the one or more surgical tools via the electronic control system to operate the one or more surgical tools. The one or more portable handheld controllers can provide said one or more control signals from a plurality of locations of an operating arena, allowing a surgeon to be mobile during a surgical procedure and to remotely operate the one or more surgical tools from different locations of the operating arena.

Abstract: Disclosed techniques can generate content object summaries. Content of a content object can be parsed into a set of word groups. For each word group, at least one topic to which the word group pertains can be identified and it can be determined, via a user model, at least one weight of the plurality of weights corresponding to the topic(s). For each word group, a score can be determined for the word group based on the weight(s). A subset of the set of word groups can be selected based on the scores for the word group. A summary of the content object can be generated that includes the subset but that does not include one or more other word groups in the set of word groups that are not in the subset. At least part of the summary of the content object can be output.

Abstract: Provided are systems, computer-implemented methods, and computer-program products for a multi-lingual device, capable of receiving verbal input in multiple languages, and further capable of providing conversational responses in multiple languages. In various implementations, the multi-lingual device includes an automatic speech recognition engine capable of receiving verbal input in a first natural language and providing a textual representation of the input and a confidence value for the recognition. The multi-lingual device can also include a machine translation engine, capable of translating textual input from the first natural language into a second natural language. The machine translation engine can output a confidence value for the translation. The multi-lingual device can further include natural language processing, capable of translating from the second natural language to a computer-based language.

Abstract: A vehicle includes a main body, a transmission having a plurality of gear stages, a data storage, and an electronic controller. The data storage stores vehicle running condition data that includes at least one of a prescribed vehicle running condition, a sensed vehicle running condition, or a transmitted vehicle running condition. The electronic controller determines a gear shift of the gear stages of the transmission based on a future vehicle speed that has been estimated based on the vehicle running condition data.

Abstract: Zero-shot content detection includes building/training a semantic space by embedding word-based document descriptions of a plurality of documents into a multi-dimensional space using a semantic embedding technique; detecting a plurality of features in the multimodal content by applying feature detection algorithms to the multimodal content; determining respective word-based concept descriptions for concepts identified in the multimodal content using the detected features; embedding the respective word-based concept descriptions into the semantic space; and in response to a content detection action, (i) embedding/mapping words representative of the content detection action into the semantic space, (ii) automatically determining, without the use of training examples, concepts in the semantic space relevant to the content detection action based on the embedded words, and (iii) identifying portions of the multimodal content responsive to the content detection action based on the concepts in the semantic space det

Abstract: Techniques for verifying identity of a human subject to an identification document are described. In some examples, a computing device may be connected to least two cameras oriented such that a first field of view is a substantially opposite direction from a second field of view. The device may receive images from the first camera that include a human subject. Second images from the second camera may include images of an ID document with a photograph of the human subject. The device may process the first images along with the respective, corresponding second images to determine respective 3D locations for at least one of cameras at the respective times the images were captured. Based on the sequence of 3D locations, along with the first images and the second images, the device may determine whether the human subject is a valid human subject.

Abstract: A minimal antigen delivery system consists essentially of PEGylated stealth liposomes loaded with an immunogenic human leukocyte antigen (HLA) class restricted peptide and surface modified with a cell targeting peptide which mediates binding and internalization of the liposomes into a target cell.

Abstract: This disclosure concerns the discovery of the use of fenoterol and (R,R)- and (R,S)-fenoterol analogues for the treatment of a tumor expressing a ?2-adrenergic receptor, such as a primary brain tumor, including a glioblastoma or astrocytoma expressing a ?2-adrenergic receptor. In one example, the method includes administering to a subject a therapeutically effective amount of fenoterol, a specific fenoterol analogue or a combination thereof to reduce one or more symptoms associated with the tumor, thereby treating the tumor in the subject.

Abstract: Techniques are disclosed for improving navigation accuracy for a mobile platform. In one example, a navigation system comprises an image sensor that generates a plurality of images, each image comprising one or more features. A computation engine executing on one or more processors of the navigation system processes each image of the plurality of images to determine a semantic class of each feature of the one or more features of the image. The computation engine determines, for each feature of the one or more features of each image and based on the semantic class of the feature, whether to include the feature as a constraint in a navigation inference engine. The computation engine generates, based at least on features of the one or more features included as constraints in the navigation inference engine, navigation information. The computation engine outputs the navigation information to improve navigation accuracy for the mobile platform.

Abstract: In general, the disclosure describes techniques for personalizing a meeting summary according to the relevance of different meeting items within a meeting to different users. In some examples, a computing system for automatically providing personalized summaries of meetings comprises a memory configured to store information describing a meeting; and processing circuitry configured to receive a plurality of meeting item summaries of respective meeting items included in the transcript of the meeting; determine, by applying a model of meeting item relevance to the meeting item summaries, a corresponding relevance to a user of each of the meeting item summaries; and output respective indications of relevance to the user for one or more of the meeting item summaries to provide a personalized summary of the meeting to the user.

Abstract: In general, the disclosure describes various aspects of techniques for attention-based explanations for artificial intelligence behavior. A device comprising a memory and a computation engine executing a processor may be configured to perform the techniques. The memory may store the artificial intelligence model and the image. The computation engine may receive a query regarding the image, and execute the artificial intelligence model to analyze the image in order to output the result to the query. The artificial intelligence model may, when analyzing the image to output the result, segment the image into hierarchically arranged semantic areas in which objects in the image are segmented into parts, determine, based on the query, an attention mask for the areas, update, based on the attention mask, the image to visually identify which of the areas formed a basis for the result, and output the updated image.

Abstract: A flexible exosuit includes rigid and flexible elements configured to couple forces to a body of a wearer. Further, the flexible exosuit includes flexible linear actuators and clutched compliance elements to apply and/or modulate forces and/or compliances between segments of the body of the wearer. The flexible exosuit further includes electronic controllers, power sources and sensors. The flexible exosuit can be configured to apply forces to the body of the wearer to enable a variety of applications. In some examples, the flexible exosuit can be configured to augment the physical strength or endurance of the wearer. In some examples, the flexible exosuit can be configured to train the wearer to perform certain physical tasks. In some examples, the flexible exosuit can be configured to record physical activities of the wearer.

Abstract: Methods and systems of using machine learning to create a trusted model that improves the operation of a computer system controller are provided herein. In some embodiments, a machine learning method includes training a model using input data, extracting the model, and determining whether the model satisfies the trust-related constraints. If the model does not satisfy the trust-related constraint, modifying at least one of: the model using one or more model repair algorithms, the input data using one or more data repair algorithms, or a reward function of the model using one or more reward repair algorithms, and re-training the model using at least one of the modified model, the modified input data, or the modified reward function. If the model satisfies the trust-related constraints, providing the model as a trusted model that enables a computer system controller to perform system actions within predetermined guarantees.

Abstract: An example system includes a vehicle, a robot, and a controller. The vehicle may include an accelerator operator and a steering operator. The robot may include as accelerator actuator configured to operate the accelerator operator, and a steering actuator configured to operate-the steering operator. The controller is configured to: in response to an accelerator command, send a first signal to the accelerator actuator to operate the accelerator operator of the vehicle, and in response to a steering command, send a second, signal to the steering actuator to steer the vehicle.

Abstract: The present inventors have developed antigen-binding reagents and antigen-binding conjugates that recognize a cancer-specific glycan (carbohydrate) modification on the human Periostin protein. Various in vitro and in vivo diagnostic and/or therapeutic methods using these compositions are also disclosed herein specifically for treating cancers that have amplification of the Mgat3 gene.

Abstract: Methods, computing devices, and computer-program products are provided for implementing a virtual personal assistant. In various implementations, a virtual personal assistant can be configured to receive sensory input, including at least two different types of information. The virtual personal assistant can further be configured to determine semantic information from the sensory input, and to identify a context-specific framework. The virtual personal assistant can further be configured to determine a current intent. Determining the current intent can include using the semantic information and the context-specific framework. The virtual personal assistant can further be configured to determine a current input state. Determining the current input state can include using the semantic information and one or more behavioral models. The behavioral models can include one or more interpretations of previously-provided semantic information.