Abstract: Machine learning model optimization systems and methods are disclosed. A system receives sensor data captured by one or more sensors of a vehicle during a first time period. The vehicle uses a first trained machine learning (ML) model for one or more decisions of a first decision type during the first time period. The system generates a second trained ML model at least in part by using the sensor data to train the second trained ML model. The system identifies an optimal trained ML model from a plurality of trained ML models. The plurality of trained ML models includes the first trained ML model and the second trained ML model. The system causes the vehicle to use the optimal trained ML model for one or more further decisions of the first decision type during a second time period after the first time period.

Abstract: A nasal shaper configured to be introduced into a nose of a subject, the nasal shaper including two tubes suitable for each being introduced into a nostril of the nose, connected together at their lower ends by a connecting bridge and a plate attached to the connecting bridge and extending opposite the connecting bridge relative to the tubes, the plate being configured to exert a pressure on the philtrum of the subject when the tubes are each introduced into a nostril of the nose.

Abstract: Techniques for facilitating the autonomous presentation of a self-driving vehicle are provided. In one example, a method can include a system operatively coupled to a processor, where the system: determines a feature of a self-driving vehicle based on information regarding an entity in a pending transaction; determines a task to be performed by the self-driving vehicle based on the feature; and generates an instruction for the self-driving vehicle to perform the task.

Abstract: Computerized systems and methods are provided for managing changes being made on a geographically diverse mobile telecommunications network. An intelligent application for network change management collects change requests to be implemented in a geographically disperse mobile telecommunications network. The intelligent application determines whether the changes to be implemented are for devices that high risk from a network impacting event, such as a major event or weather. The application initiates actions to suspend the changes from being implemented to the network until the devices are no longer at high risk from a network impacting event.

Abstract: A method of forming a microelectronic device comprises forming a stack structure comprising a vertically alternating sequence of insulative structures and additional insulative structures, at least some of the additional insulative structures comprising silicon nitride having a ratio of nitrogen atoms to silicon atoms greater than about 1.58:1.00, forming openings through the stack structure, and forming cell pillar structures within the openings, the cell pillar structures individually comprising a semiconductor channel material vertically extending through the stack structure. Related methods, microelectronic devices, memory devices, and electronic systems are also described.

Abstract: Described herein are, among other things, techniques for implementing a dual detail encoding scheme in a distributed display system. At the host computer, an application may render a frame for a scene at a first resolution. The host computer may generate a two-dimensional (2D) array of pixels for the frame, encode the 2D array of pixels, and transmit the encoded pixel data to the display device. The 2D array of pixels comprises, for each eye, a copy of the frame downscaled to a second resolution less than the first resolution, and a copy of a subregion of the scene. At the display device, the encoded pixel data is decoded to obtain the 2D array of pixels for the frame, which is modified by enlarging the copies of the frame so that images can be generated using the enlarged copies of the frame and the copies of the subregion with blending.

Abstract: Optical positional tracking systems that may be used in virtual reality (VR)/augmented reality (AR) applications are described. Exemplary implementations comprise one or more receivers and one or more transmitters. Exemplary transmitters contain two orthogonal rotors that each emit a fan-shaped laser beam. Each beam is swept as the rotors are spun at constant speed. Exemplary optical receivers can be relatively small, and mounted at convenient locations on the VR display. These receivers consist of small optical detectors that may be mounted on head-mounted VR displays. Exemplary systems determine position by measuring the time at which each swept beam crosses each receiver/detector.

Abstract: A method of forming a microelectronic device comprises forming a stack structure comprising a vertically alternating sequence of insulative structures and additional insulative structures, at least some of the additional insulative structures comprising silicon nitride having a ratio of nitrogen atoms to silicon atoms greater than about 1.58:1.00, forming openings through the stack structure, and forming cell pillar structures within the openings, the cell pillar structures individually comprising a semiconductor channel material vertically extending through the stack structure. Related methods, microelectronic devices, memory devices, and electronic systems are also described.

Abstract: Some embodiments may include a touchscreen or other user input interface to select part of a displayed map and one or more processors coupled to the touchscreen or other user input interface. The one or more processors may be configured to setup an autonomous operating zone for a working machine based on a user selection from a displayed map. The working machine may monitor its current location with respect to the autonomous operation zone, and may de-actuate at least one of its actuator(s) or send a new actuation signal to its actuator(s) to change an operation of at least one motorized device of the working machine. Other embodiments may be disclosed and/or claimed.

Abstract: An integrated, modular system for direct air capture (DAC) and electro-microbial production (EMP) for bioelectrochemical conversion of CO2 comprises (a) a solid absorbent configured to directly capture CO2 from air; (b) a first bioreactor configured to receive enriched and purified CO2 from the absorbent and convert the CO2 to an upgradeable organic carbon intermediate by an autotrophic microorganism, wherein the autotrophic organism derives energy from oxidation of electrochemically-generated reducing equivalents; and (c) a second bioreactor configured to receive the organic carbon intermediate from the first bioreactor for use as a feedstock by a metabolically engineered microorganism to generate a value added product.

Abstract: A queue management system including a user data retrieving circuit configured to retrieve a user property of the user and a queue managing circuit configured to model the plurality of users as a non-Newtonian fluid, to analyze the user properties to assign each user an anisometric polarity parameter, and to solve an objective function for a queue property to cause the plurality of users to collectively act as the non-Newtonian fluid in the queue.

Abstract: A microelectronic device includes tiers of alternating dielectric and conductive materials, a cap oxide material vertically adjacent to the tiers, and pillars extending vertically through the tiers. The cap oxide material is formulated to exhibit a different etch rate relative to an etch rate of the oxide material of the tiers. Additional microelectronic devices, microelectronic systems, and methods of forming a microelectronic device are also disclosed.

Abstract: Machine learning model optimization systems and methods are disclosed. A system receives sensor data captured by one or more sensors of a vehicle during a first time period. The vehicle uses a first trained machine learning (ML) model for one or more decisions of a first decision type during the first time period. The system generates a second trained ML model at least in part by using the sensor data to train the second trained ML model. The system identifies an optimal trained ML model from a plurality of trained ML models. The plurality of trained ML models includes the first trained ML model and the second trained ML model. The system causes the vehicle to use the optimal trained ML model for one or more further decisions of the first decision type during a second time period after the first time period.

Abstract: A modular and scalable system configured to generate a topology platform for training, entertainment, gaming, living, working, acting, viewing, etc. The platform can have a plurality of movable columns that are extendable and retractable. In some embodiments, some of the columns can be retracted to allow users to walk on the columns while other columns can be extended to form a barrier (e.g., wall) or form an inanimate object (e.g., table). The system may be used in furniture to provide topologies for highly customized support structure designs to accommodate varying body shapes, sizes, and individual comfort levels. The system may be used to generate topologies for greenscreen film stages to provide post-production editing, compositing with actors, stuntmen, and video, etc. The platform can be configurable and re-configurable to change the objects and features of the topology and represent different or dynamic environments at any time.

Abstract: Computerized systems and methods are provided for managing changes being made on a geographically diverse mobile telecommunications network. An intelligent application for network change management collects change requests to be implemented in a geographically disperse mobile telecommunications network. The intelligent application determines whether the changes to be implemented are for devices that high risk from a network impacting event, such as a major event or weather. The application initiates actions to suspend the changes from being implemented to the network until the devices are no longer at high risk from a network impacting event.

Abstract: Techniques for facilitating the autonomous presentation of a self-driving vehicle are provided. In one example, a method can include a system operatively coupled to a processor, where the system: determines a feature of a self-driving vehicle based on information regarding an entity in a pending transaction; determines a task to be performed by the self-driving vehicle based on the feature; and generates an instruction for the self-driving vehicle to perform the task.

Abstract: Disclosed is a medical device comprising a porous structure, wherein a configuration of the porous structure varies in dependence on a load applied to the porous structure, such that the porous structure has a first configuration when the load is of a first magnitude, and has a second configuration when the load is of a second magnitude greater than the first magnitude. The porous structure comprises a first surface portion and a second surface portion. The first surface portion is disengaged from the second surface portion when the porous structure has the first configuration, and is engaged with the second surface portion when the porous structure has the second configuration.

Abstract: A medical device comprising a substantially flexible porous structure. The porous structure comprises a plurality of interlocking units. Each of the plurality of interlocking units comprises a body and at least one arm. The plurality of interlocking units is configured to have space between adjacent interlocking units when the porous structure is in a neutral configuration. The plurality of interlocking units is configured to contact the respective body and arm of adjacent interlocking units when a compressive force is applied to the porous structure, thereby restricting compression of the porous structure. The plurality of interlocking units is configured to contact the respective arms of adjacent interlocking units when an extension force is applied to the porous structure, thereby restricting extension of the porous structure.

Abstract: Described are techniques for differentiating humans from bots. The techniques including a computer-implemented method comprising presenting a motion-based challenge-response instruction to a user via a user interface of a first device of a plurality of devices associated with the user and communicatively coupled to one another by a network, where the motion-based challenge-response instruction describes at least one motion that is performable by the user and detectable by at least one of the plurality of devices, and where the motion-based challenge-response instruction is configured to differentiate humans from bots. The method further comprises determining that device data from one or more of the plurality of devices matches the at least one motion. The method further comprises authenticating the first device in response to determining that the device data matches the at least one motion, where authenticating the first device indicates that the user is a human.

Abstract: The disclosure relates generally to techniques for determining pupil location of a display device's user via imaging sensors on the display device, and using that information to verify and/or correct positioning of the display device or its internal components. The display device may be a head-mounted display (“HMD”) device with display panels separated from a wearer's eyes via intervening lenses, with the sensors including optical flow sensor integrated circuits mounted on or near at least one of the display panels to capture images of the wearer's eye locations through the lenses, and with the correction to the positioning including modifications to the alignment or other positioning of the HMD device on the wearer user's head and/or its internal components within the HMD device (e.g., based on automated control of motors on the HMD device) to reflect a target alignment of the wearer's eyes relative to displayed information.