Abstract: Provided are a computer-implemented method, a system, and a computer program product. The method comprises extracting features from a plurality of base models in an ensemble model. The plurality of base models are configured to provide respective prediction results. The ensemble model is configured to provide an overall prediction result from the prediction results of the plurality of base models. The features are associated with time performance of the base models. The method further comprises clustering the plurality of base models into a plurality of clusters based on the extracted features. The method further comprises assigning the plurality of base models to a plurality of parallel computation units based on the plurality of clusters.

Abstract: Disclosed are a computer-implemented method, a system and a computer program product for model exploration. Model feature importance of each model of a plurality of models can be obtained, the plurality of models can be grouped into a plurality of model clusters based on the model feature importance of each model, and the model feature importance can be presented by box-plot or confidence interval.

Abstract: The present disclosure is directed to an apparatus. The apparatus can include an actuator having a plurality of openings. Each of the plurality of openings can couple with a thermal management loop. The thermal management loop can couple with a battery module cold plate. The actuator can select, based on a parameter, one fluid channel of the thermal management loop to control fluid distribution through the plurality of openings and through the thermal management loop to regulate a temperature of the battery module.

Abstract: The present disclosure relates to privacy protection in a search process. According to a method, a target emotion vector is extracted from a search interaction, the target emotion vector representing emotional information in the search interaction. Respective emotion distances between the target emotion vector and respective emotion vectors associated with a plurality of text clusters are determined. The plurality of text clusters is clustered from a dictionary of text elements. A first number of text clusters are selected from the plurality of text clusters based on the determined respective emotion distances. The first number of text clusters have emotion distances larger than at least one unselected text cluster among the plurality of text clusters. A plurality of confused search interactions are constructed for the search interaction based on the first number of text clusters, and the plurality of confused search interactions are performed.

Abstract: A treatment planning prediction method to predict a Dose-Volume Histogram (DVH) or Dose Distribution (DD) for patient data using a machine-learning computer framework is provided with the key inclusion of a Planning Target Volume (PTV) only treatment plan in the framework. A dosimetric parameter is used as an additional parameter to the framework and which is obtained from a prediction of the PTV-only treatment plan. The method outputs a Dose-Volume Histogram and/or a Dose Distribution for the patient including the prediction of the PTV-only treatment plan. The method alleviates the complicated process of quantifying anatomical features and harnesses directly the inherent correlation between the PTV-only plan and the clinical plan in the dose domain. The method provides a more robust and efficient solution to the important DVHs prediction problem in treatment planning and plan quality assurance.

Abstract: A fingerprint detection apparatus includes a plurality of fingerprint detection units distributed in an array or disposed alternately, and each of the plurality of fingerprint detection units includes: one micro-lens configured to be disposed under the display screen; a plurality of optical sensing pixels, a center position of a photosensitive region of each of the plurality of optical sensing pixels being offset relative to a center position of the same optical sensing pixel; and at least one light shielding layer disposed between the one micro-lens and the plurality of optical sensing pixels, each of the at least one light shielding layer being provided with a hole corresponding to the plurality of optical sensing pixels. By offsetting photosensitive regions of the optical sensing pixels, the thickness of an optical fingerprint module can be reduced on the basis of improving a fingerprint identification effect on a dry finger.

Abstract: A fingerprint detection apparatus and an electronic device are provided. The fingerprint detection apparatus is applied under a display screen to implement under-screen optical fingerprint identification, including: a fingerprint sensor; the fingerprint sensor includes: N groups of sub-pixels, where N is a positive integer greater than 1; and different groups of sub-pixels in the N groups of sub-pixels respectively receive light signals with different characteristics in light signals returned by reflection or scattering via a finger, and light signals received by each group of sub-pixels in the N groups of sub-pixels are used to form one pixel in a fingerprint image of the finger.

Abstract: Examples described herein provide a computer-implemented method that includes determining a kernel width for the machine learning model. The method further includes building a local interpretable linear model using the kernel width. The method further includes computing a contribution and confidence for a feature of the local interpretable linear model. The method further includes updating the local interpretable linear model to generate a final model and computing an overall confidence for the final model.

Abstract: A method, computer program, and computer system are provided for training a machine learning model. A feature associated with training data derived from a dataset is identified. A machine learning model is generated based on the training data. At least a portion of the training data associated with maximizing an importance value associated with the identified feature is selected. The importance value corresponds to a need associated with the machine learning model. One or more weight values is assigned to the selected portion of the training data. The machine learning model is updated based on the assigned weight values.

Abstract: Illustrative examples of an apparatus may include a quenching chamber configured to direct a heat flow from a battery cell compartment to a vent. The quenching chamber may be configured to extinguish a flame present in the heat flow before the heat flow reaches the vent. In other examples, a method includes receiving a heat flow from a battery cell compartment comprising a plurality of battery cells at a quenching chamber. The method further comprises directing the thermal runaway heat flow through the quenching chamber to a vent.

Abstract: A method, computer system, and a computer program product for performing an interactive outcome analysis is provided. The present invention may include generating, by a computer, a first estimation outcome from a first plurality of input conditions. The present invention may include generating, by the computer, a parallel estimation outcome from a second plurality of input conditions, wherein at least one of said input conditions in said first plurality of input conditions is different from any of said second plurality of input conditions. The present invention may include selecting, by the computer, either said first or said parallel estimation outcome by analyzing said outcomes with one another and with a target goal outcome.

Abstract: A multiview backlight, multiview display, and method employ arrays of active emitters arranged in columns to provide directional light beams having directions corresponding to view directions of a multiview display. The multiview backlight includes a first array of active emitters arranged in columns and configured to provide a first plurality of directional light beams and a second array of active emitters arranged in columns and configured to provide a second plurality of directional light beams. Columns of the second active emitter array are disposed between and alternate with the columns of the first active emitter array. The multiview display further includes an array of light valves configured to modulate the directional light beams to display a multiview image. The view directions of the multiview display are in a plane that crosses the columns of first and second active emitter arrays.

Abstract: A multiview backlight, multiview display, and method employ arrays of active emitters and a diffuser to provide an effective active emitter from emitted light from each active emitter. The multiview backlight includes an array of active emitters and a diffuser configured to provide effective active emitters having a predetermined size that is between one quarter and two times a size of a light valve of the multiview display. The multiview display includes an array of effective active emitters including active emitters and a diffuser configured to provide output emitted light. The multiview display further includes an array of light valves configured to modulate output emitted light to provide a displayed image. A spacing between effective active emitters is an integer multiple of a spacing between light valves of the multiview display.

Abstract: A method of clustering data generated by an unknown model, the method including accessing the data, wherein the data includes a prediction target and a confidence, extracting a data group with high prediction confidence from the data, wherein the data group comprises a plurality of data cases, and where each of the data cases is described by a plurality of predictors, identifying high rank predictors of each the data cases in the data group, transforming the data group into a transformed data group including only the high rank predictors for each of the data cases, wherein the high rank predictors are ranked within each of the data cases included in the transformed data group, clustering the transformed data group to generate clusters, and profiling the clusters to extract an insight about the unknown model.

Abstract: The present disclosure relates to crystalline forms of benzamide compound and processes for preparing the same. Provided are crystalline Form J and an amorphous form of a compound of formula (I), a process for preparing the same, and use thereof. The crystalline Form J has the advantages in at least one of solubility, melting point, stability, dissolution, hygroscopicity, adhesion, fluidity, biological effectiveness, processing performance, purification, formulation production, safety, and the like, which provides a new and better choice for the preparation of pharmaceutical formulations containing the compound of formula (I), and has a very important significance for the drug development.

Abstract: A computer-implemented method, system and computer program product for generating personalized recommendations to address a target problem. A machine learning prediction model directed to a target problem for an individual is built with historical data. After receiving data about the individual, a prediction for the individual is obtained in connection with the target problem by the built model using the received data about the individual. Key predictors (e.g., parameters) and their weight for the individual are generated using the prediction by an explanation model. Record(s) are identified from the historical data by performing similarity analysis of the historical data using the key predictors and their weight. Such records provide a population closely related to the individual with respect to the target problem. These records are then analyzed and recommendations are provided to a user to solve the target problem for the individual based on the analysis of the identified record(s).

Abstract: An approach is provided in which the approach generates a parallel chart based on multiple records that includes a set of variable values corresponding to a set of variables. To generate the parallel chart, the approach arranges the set of variables on the parallel chart in a variable order based on at least one variable arrangement rule. The approach arranges an initial variable value order for each one of the set of variables, and computes a lucidity score based on the variable order and the initial variable value order of each of the set of variables. The lucidity score is a measurement of the clarity of the parallel chart. The approach adjusts the variable value order of at least one of the set of variables to increase the lucidity score and optimizes the clarity of the parallel chart based on the adjusted variable value order.

Abstract: A collimated backlight and an electronic display employ a light guide having angle-preserving scattering feature and an absorption collimator. The angle-preserving scattering feature is configured to scatter a portion of guided light out of the light guide as emitted light. The absorption collimator includes an absorption element and is configured to convert light provided by a light source into collimated light to be guided as the guided light. The electronic display includes an array of light valves and may be configured as a multiview display or a privacy display.

Abstract: Various disclosed embodiments include illustrative control units, systems, and methods. In an illustrative embodiment, a control unit includes a processor and non-transitory computer-readable media that stores computer-executable instructions configured to cause the processor to determine a start charging time, receive a battery temperature value, and instruct the thermal control system to start heating the battery responsive to the start charging time and the battery temperature value in an energy saving way.

Abstract: A computer implemented technique including: splitting data of a historical time series data set into subsets; updating a time series model by backwards data selection to obtain an interim version of the time series model; exploring pattern changes in the new data to obtain new predictors of pattern change; and updating the interim version of the time series model by applying the new predictors of pattern change to obtain an updated version of the time series model.