Abstract: To automate time series forecasting machine learning pipeline generation, a data allocation size of time series data may be determined based on one or more characteristics of a time series data set. The time series data may be allocated for use by candidate machine learning pipelines based on the data allocation size. Features for the time series data may be determined and cached by the candidate machine learning pipelines. Predictions of each of the candidate machine learning pipelines using at least the one or more features may be evaluated. A ranked list of machine learning pipelines may be automatically generated from the candidate machine learning pipelines for time series forecasting based upon evaluating predictions of each of the one or more candidate machine learning pipelines.

Abstract: Disclosed herein is a flexible sensing interface, comprising: a sensor, comprising: a core; an inner electrode in the form of a conductive material in contact with the core; an inner dielectric material substantially encasing the inner electrode; an outer electrode in the form of a conductive material in contact with the inner dielectric material and in electrical communication with the inner electrode; and an outer dielectric material substantially encasing the outer electrode; wherein the inner dielectric material and the outer dielectric material comprise an elastic material. Also disclosed herein are systems and methods for making and using the same.

Abstract: The disclosure relates to nanoparticle drug conjugates (NDC) that comprise ultrasmall nanoparticles, folate receptor (FR) targeting ligands, and linker-drug conjugates, and methods of making and using them to treat cancer.

Abstract: Various embodiments described herein provide a variable focus lens assembly. Some embodiments are designed to enable repositioning of one or more components, such as a lens barrel assembly, to adjust the focus of the variable focus lens assembly. Some example variable focus lens assemblies include a module base housing a lens barrel assembly having a pair of positioning magnets, a pair of positioning coil assemblies associated with the positioning magnets, and at least one pair of bearing balls movably supporting the lens barrel assembly. The positioning coil assemblies together with the positioning magnets are configured to exert various magnetic fields to reposition the lens barrel assembly. Further embodiments are provided for imaging apparatus including at least one variable focus lens assembly described herein.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using embedded function with a deep network. One of the methods includes receiving an input comprising a plurality of features, wherein each of the features is of a different feature type; processing each of the features using a respective embedding function to generate one or more numeric values, wherein each of the embedding functions operates independently of each other embedding function, and wherein each of the embedding functions is used for features of a respective feature type; processing the numeric values using a deep network to generate a first alternative representation of the input, wherein the deep network is a machine learning model composed of a plurality of levels of non-linear operations; and processing the first alternative representation of the input using a logistic regression classifier to predict a label for the input.

Abstract: Building infrastructure and robot coordination methods and systems are disclosed herein. An example method can include preregistering an autonomous vehicle with a building controller of a building to inform the building controller of arrival of the autonomous vehicle and a delivery request for a package, the delivery request specifying a recipient of the package, the recipient having a location in the building. The method can include determining completion of a validation routine with the autonomous vehicle, assigning a task management plan to the autonomous vehicle, the task management plan determining how the autonomous vehicle navigates through the building to deliver the package to the location, and tracking the autonomous vehicle through the building.

Abstract: An apparatus to facilitate single precision support for systolic pipeline in a graphics environment is disclosed. The apparatus includes a processor comprising systolic array hardware including a plurality of data processing units, wherein the systolic array hardware is to: receive data for performance of a matrix multiplication operation in a first precision format; convert an original value of the data into two split values with a second precision format having a lower precision than the first precision format; perform the matrix multiplication operation using the two split values in the second precision format, the matrix multiplication operation comprising a split-term operation that utilizes two passes through the systolic array hardware with feedback wiring and local reduction; and generate an emulated result for the matrix multiplication operation in the first precision format.

Abstract: Some implementations of the disclosure relate to an imaging system including one or more image sensors and a Z-stage. The imaging system is configured to perform operations including: capturing, using the one or more image sensors, a first image of a first pair of spots projected at a first sample location of a sample; determining whether or not the first image of the first pair of spots is valid; and when the first image is determined to be valid: obtaining, based on the first image, a current separation distance measurement of the first pair of spots; and controlling, based at least on the current separation distance measurement, the Z-stage to focus the imaging system at the first sample location.

Abstract: An apparatus to facilitate hardware enhancements for double precision systolic support is disclosed. The apparatus includes matrix acceleration hardware having double-precision (DP) matrix multiplication circuitry including a multiplier circuits to multiply pairs of input source operands in a DP floating-point format; adders to receive multiplier outputs from the multiplier circuits and accumulate the multiplier outputs in a high precision intermediate format; an accumulator circuit to accumulate adder outputs from the adders with at least one of a third global source operand on a first pass of the DP matrix multiplication circuitry or an intermediate result from the first pass on a second pass of the DP matrix multiplication circuitry, wherein the accumulator circuit to generate an accumulator output in the high precision intermediate format; and a down conversion and rounding circuit to down convert and round an output of the second pass as final result in the DP floating-point format.

Abstract: Systems and methods for obtaining an external content item from an online service for presentation on the client device is presented. On the client device, a condition is detected in the execution context of an executing application, the condition corresponding to the presentation of an external content item by the application. A request is made to the online service for content items. In response, content information is received, where the content information identifies potential content items. According to various embodiments, initial scores are associated with each potential content item in the content information. At the client device, a completed score is generated for each content item according to local information on the client device and the corresponding initial score. An external content item is selected for presentation on the client device from the potential content items according to, at least in part, the completed scores.

Abstract: A location-aware electronic device is provided. The electronic device trains feature extraction layers, reconstruction layers, and classification layers. The training may be based on a reconstruction loss and/or a clustering loss. The electronic device processes a fingerprint to obtain an augmented fingerprint using randomization based on statistics of the fingerprint. The feature extraction layers provide feature data to both the reconstruction layers and the classification layers. The classification layers operate on the codes to obtain an estimated location label. An application processor operates on the estimated location label to provide a location-aware application result to a person.

Abstract: Systems and methods are provided for centralized validation of potential network calls, such as calls proposing database transactions, on a distributed system. The distributed system may include multiple systems that apply independent criteria for validating proposals, which criteria may not be available external to the individual systems. Moreover, the systems may lack an ability to validate proposals prior to submitting such proposals for commitment. A centralized network call parameter validation system as disclosed herein may validate potential network calls with high confidence by applying probability models of data pattern and hashing digit checksum to potential network call parameter values, which models are generated based on statistical analysis of historical network call values.

Abstract: A system for predicting and summarizing medical events from electronic health records includes a computer memory storing aggregated electronic health records from a multitude of patients of diverse age, health conditions, and demographics including medications, laboratory values, diagnoses, vital signs, and medical notes. The aggregated electronic health records are converted into a single standardized data structure format and ordered arrangement per patient, e.g., into a chronological order. A computer (or computer system) executes one or more deep learning models trained on the aggregated health records to predict one or more future clinical events and summarize pertinent past medical events related to the predicted events on an input electronic health record of a patient having the standardized data structure format and ordered into a chronological order.

Abstract: An approach to precision additive fabrication uses jetting of cationic compositions in conjunction with a non-contact (e.g., optical) feedback approach. By not requiring contact to control the surface geometry of the object being manufactured, the approach is tolerant of the relative slow curing of the cationic composition, while maintaining the benefit of control of the deposition processes according to feedback during the fabrication processes. This approach provides a way to manufacture precision objects and benefit from material properties of the fabricated objects, for example, with isotropic properties, which may be at least partially a result of the slow curing, and flexible structures, which may not be attainable using conventional jetted acrylates.

Abstract: A virtual railroad of vehicles is disclosed. In one aspect of the disclosure, a system includes one or more passenger vehicles of a peloton, and a first engine vehicle of the peloton. The first engine vehicle communicatively connected to the one or more passenger vehicles, wherein the first engine vehicle comprises: a processor communicatively connected to a memory and is configured to receive status information of the one or more passenger vehicles, determine, based on the received status information, a set of current values for a set of vehicle attributes for each of the one or more passenger vehicles, and adjust, based on the set of current values for the set of vehicle attributes, a position of a corresponding passenger vehicle of the one or more passenger vehicles.

Abstract: Hybrid use of dual policies is provided to improve a communication system. In a multiple access scenario, when an inactive user equipment (UE) transitions to an active state, it may be become a burden to a radio cell on which it was previously camping. In some embodiments, hybrid load balancing is provided using a hierarchical machine learning paradigm based on reinforcement learning in which an LSTM generates a goal for one policy influencing cell reselection so that another policy influencing handover over active UEs can be assisted. The communication system as influenced by the policies is modeled as a Markov decision process (MDP). The policies controlling the active UEs and inactive UEs are coupled, and measureable system characteristics are improved. In some embodiments, policy actions depend at least in part on energy saving.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for computing numeric representations of words. One of the methods includes obtaining a set of training data, wherein the set of training data comprises sequences of words; training a classifier and an embedding function on the set of training data, wherein training the embedding function comprises obtained trained values of the embedding function parameters; processing each word in the vocabulary using the embedding function in accordance with the trained values of the embedding function parameters to generate a respective numerical representation of each word in the vocabulary in the high-dimensional space; and associating each word in the vocabulary with the respective numeric representation of the word in the high-dimensional space.

Abstract: A real time Taq-Man PCR assay for detecting multiple serotypes of human papillomavirus (HPV) wherein the number of serotypes detected exceeds the number of colorimetric channels for detection. A biological sample is combined with three oligonucleotide primer/probe sets such that the probes and primers anneal to a target sequence. Each primer/probe set is at least preferential for a specific serotype of an organism. The first and second primer/probe sets are degenerate with respect to each other. The third primer/probe set is not degenerate with respect to the first and second primer/probe sets and discriminates for a third serotype. The third primer/probe set has a signal moiety that emits signal at a wavelength that is the same or different from the wavelength emitted by the signal moiety of the degenerate primer/probe set probes. The target sequences, if present, are amplified and detected.

Abstract: The present invention provides compositions and methods for stimulating an immune response using cationic lipids alone or in combination with antigens.

Abstract: A method for linking a natural product and gene cluster is disclosed. In some embodiments, monomers of natural products are predicted from a gene sequence. In other embodiments, monomers of natural products are predicted from a chemical structure of a natural product. In another embodiment, monomers predicted from gene sequences are aligned with monomers predicted from chemical structures.