Abstract: Described are techniques for machine learning library management. The techniques include generating a table including a plurality of machine learning libraries and their current versions that are used in a deployed machine learning platform (MLP) instance, a first available version upgrade for a first machine learning library of the plurality of machine learning libraries, a security indication associated with the first available version upgrade relative to a current version implemented by the first machine learning library, and a compatibility indication between the first available version upgrade and the current version of the first machine learning library. The techniques further include generating a recommendation related to upgrading the first machine learning library based on the security indication and the compatibility indication.

Abstract: Described are techniques for machine learning library management. The techniques include generating a table including a plurality of machine learning libraries and their current versions that are used in a deployed machine learning platform (MLP) instance, a first available version upgrade for a first machine learning library of the plurality of machine learning libraries, a security indication associated with the first available version upgrade relative to a current version implemented by the first machine learning library, and a compatibility indication between the first available version upgrade and the current version of the first machine learning library. The techniques further include generating a recommendation related to upgrading the first machine learning library based on the security indication and the compatibility indication.

Abstract: A computer-implemented system, platform, programing product, and/or method for improving transformation selection in an ensemble machine learning (ML) model that includes: providing all base ML models of the ensemble ML model; identifying all of a plurality of Derived Fields in all the base ML models; performing a Derived Field run prediction analysis for all the Derived Fields; computing the Derived Field Importance Weight for Field (DFIW4F) and the Derived Field Importance Weight for Model (DFIW4M) for all the Derived Fields; clustering all the Derived Fields into a plurality of Derived Field clusters, wherein each Derived Field cluster is based upon the DFIW4M and the DFIW4F for the Derived Field; sorting all the Derived Field clusters by best cluster based upon DFIW4M and DFIW4F; and running the base ML models based upon the Derived Fields in the best Derived Field cluster until sufficient base ML models have been run.

Abstract: Some embodiments of the disclosure provide a cathode circulation system of a fuel cell connected to a power generation unit of the fuel cell. The cathode circulation system includes a first gas supply tank for providing an inert gas, a second gas supply tank for providing a reaction gas, a mixing tank connected to the first gas supply tank and the second gas supply tank for mixing the inert gas and the reaction gas, a gas-liquid separator connected to the power generation unit, and at least one cathode gas pump provided between the mixing tank and the gas-liquid separator and between the mixing tank and the power generation unit.

Abstract: Some embodiments of the disclosures provide an anode recovery system of a fuel cell. The anode recovery system includes a gas supply unit connected to a power generation unit, a gas-liquid separator connected to the power generation unit and a first anode gas recovery control component; and a storage tank connected to the gas-liquid separator and a cathode recovery system. The gas supply unit is configured to provide anode gas to the power generation unit. A first part of unreacted anode gas from the power generation unit mixes with the anode gas and flows back to the power generation unit sequentially via the gas-liquid separator and the first anode gas recovery control component. A second part of the unreacted anode gas and generated water are discharged from the power generation unit to the storage tank and is further discharged to the cathode recovery system.

Abstract: Techniques for optimizing machine learning models are provided. A first model in a first format is received. A second model is generated by applying one or more optimization techniques to the first model; the second model is an optimized version of the first model. The second model is converted into a common intermediate format. The second model is converted into binary data representing the second model. The binary data representing the second model is outputted.

Abstract: A system and related method identify a weakness of a workflow in a complex system. The method collects runtime data about the complex system. The complex system comprises a plurality of subcomponents, and the method identifies an abnormal operation in the complex system. The method constructs a multi-dimensional cause-and-effect relation matrix among the plurality of subcomponents, and filters one or more related operations using the multi-dimensional cause-and-effect relation matrix.

Abstract: A computer-implemented method, system and computer program product for accurately identifying an execution time of a performance test. Network latency data is grouped into clustered groups of network latency data. Furthermore, the performance test execution times for the same group of performance tests run in the local and remote cluster environments are obtained. The test execution times impacted by network latency (compensation times) are then determined based on such obtained performance test execution times in the local and remote cluster environments. Such compensation times are then grouped into clustered groups of compensation times. A regression model is built to predict a performance test execution time impacted by network latency (compensation time) using the clustered groups of network latency data and compensation times.

Abstract: One or more computer processors calculate a cache prediction for a received inference request within an inference cache structured as a self-learning tree, wherein the inference request comprises a set of input values. The one or more computer processors responsive to the retrieved cache prediction exceeding a cache prediction threshold, transmit the cache prediction. The one or more computer processors parallel compute a model prediction for the received inference request utilizing a trained model. The one or more computer processors responsive to the retrieved model prediction exceeding a model prediction threshold, convert the trained model into a tree structure. The one or more computer processors update the inference cache with the converted train model. The one or more computer processors transmit the model prediction.

Abstract: Embodiments of the present disclosure relate to methods, systems, and computer program products for optimized prediction of a tree ensemble. According to a method, an input request is received, which indicates a plurality of input values for a plurality of variables associated with a tree ensemble. A plurality of target transformed intervals, into which the plurality of input values fall respectively, are determined by matching the plurality of input values with a plurality of sets of transformed intervals for the plurality of variables respectively. Respective prediction results for a plurality of tree models of the tree ensemble are determined based on the plurality of target transformed intervals and respective node hierarchies of the plurality of tree models. A tree ensemble prediction result is determined for the input request based on the determined prediction results of the plurality of tree models.

Abstract: A computer-implemented method, system and computer program product for accurately identifying an execution time of a performance test. Network latency data is grouped into clustered groups of network latency data. Furthermore, the performance test execution times for the same group of performance tests run in the local and remote cluster environments are obtained. The test execution times impacted by network latency (compensation times) are then determined based on such obtained performance test execution times in the local and remote cluster environments. Such compensation times are then grouped into clustered groups of compensation times. A regression model is built to predict a performance test execution time impacted by network latency (compensation time) using the clustered groups of network latency data and compensation times.

Abstract: Systems and computer-implemented methods select a subset of methods to generate data schemas for input data from a list of methods for generating data schemas, based on output of a regression model; generate a candidate schema for each method in the subset of methods to generate data schemas; and generate a master data schema for the input data by merging the candidate schema for each method in the subset of methods to generate data schemas, utilizing predetermined rules.

Abstract: A method for instantiating a first specific model, as one record, based on a model scoring template, clustering a plurality of models into a plurality of groups to obtain a plurality of clusters, computing a model contribution weight value for the given cluster, choosing a set of top model(s) from the plurality of models, and updating the model(s) of the set of top model(s) dynamically.

Abstract: The example embodiments disclose a system and method, a computer program product, and a computer system for improving solid-state drive performance. The example embodiments may include generating, by an affinity adapter located external to the solid-state drive, a plurality of affinities for each of a plurality of data to a respective plurality of subdivisions of data of a solid-state drive, wherein each of the plurality of data is associated with a logical block address (LBA) and each of the respective plurality of subdivisions has a physical block address (PBA).

Abstract: Embodiments of the present invention disclose a method, computer program product, and system for estimating the results of a performance test on an updated software application. A method, the method comprising receiving an updated software application, wherein the size of the updated software application is a first size and generating a plurality of small probe, wherein the size of each of the small probe data is a second size, wherein the second size is less than the first size. Conducting a first performance test on the plurality of small probe data and calculating an estimated elapsed time for a performance test on the updated software application. Conducting the performance test on the updated software application and determining if the updated software is given a PASS or FAIL for the performance test, based in part on the elapsed time of the performance test on the updated software application.

Abstract: The present invention relates to an AMOLED display, more specifically, to an AMOLED display employing an ultra-thin all-in-one substrate and an auxiliary supporting means to achieve true foldable display devices.

Abstract: In an approach to intelligently choosing transport channels across protocols by drive type, one or more transport channels on a host are detected. One or more storage drive tiers on a target are detected. The one or more transport channels are mapped to the one or more storage drive tiers, based on performance. A transfer of a data between a host and the target is completed, using each transport channel and each storage drive tier.

Abstract: The example embodiments disclose a system and method, a computer program product, and a computer system for improving solid-state drive performance. The example embodiments may include generating, by an affinity adapter located external to the solid-state drive, a plurality of affinities for each of a plurality of data to a respective plurality of subdivisions of data of a solid-state drive, wherein each of the plurality of data is associated with a logical block address (LBA) and each of the respective plurality of subdivisions has a physical block address (PBA).

Abstract: Methods that can offload data operations to a storage system are provided. One method includes performing, by a processor, a set of non-storage operations at a storage system for data stored in a set of storage devices on the storage system to generate a set of results, in which the storage system is separate from a client device that owns the data, and transmitting the result(s) to the client device that owns the data. Apparatus, systems, and computer program products that can include, perform, and/or implement the methods are also disclosed herein.