Abstract: A method for deep metric learning per hierarchical step of multi-labels and few-shot inference using the same is provided, which includes sampling a plurality of N multiplets from a learning data set that is expressed by hierarchical multi-labels; configuring the N multiplets into triplets or quadruplets so as to correspond to the number of classes in each hierarchy; performing a metric learning per hierarchical step for a deep neural network model based on the configured triplets and quadruplets; and distributing, to an embedded device, the deep neural network model from which a loss function is removed as the metric learning is completed.

Abstract: Disclosed herein are a method for performing a dilated convolution operation using an atypical kernel pattern and a dilated convolutional neural network system using the same. The method for performing a dilated convolution operation includes learning a weight matrix for a kernel of dilated convolution through deep learning, generating an atypical kernel pattern based on the learned weight matrix, and performing a dilated convolution operation on input data by applying the atypical kernel pattern to a kernel of a dilated convolutional neural network.

Abstract: Disclosed herein are an apparatus and method for machine learning based on monotonically increasing quantization resolution. The method, in which a quantization coefficient is defined as a monotonically increasing function of time, includes initially setting the monotonically increasing function of time, performing machine learning based on a quantized learning equation using the quantization coefficient defined by the monotonically increasing function of time, determining whether the quantization coefficient satisfies a predetermined condition after increasing the time, newly setting the monotonically increasing function of time when the quantization coefficient satisfies the predetermined condition, and updating the quantization coefficient using the newly set monotonically increasing function of time.

Abstract: Disclosed herein are an apparatus and method for adaptively accelerating a BLAS operation based on a GPU. The apparatus for adaptively accelerating a BLAS operation based on a GPU includes a BLAS operation acceleration unit for setting optimal OpenCL parameters using machine-learning data attribute information and OpenCL device information and for creating a kernel in a binary format by compiling kernel source code; an OpenCL execution unit for creating an OpenCL buffer for a BLAS operation using information about an OpenCL execution environment and the optimal OpenCL parameters and for accelerating machine learning in an embedded system in such a way that a GPU that is capable of accessing the created OpenCL buffer performs the BLAS operation using the kernel, and an accelerator application unit for returning the result of the BLAS operation to a machine-learning algorithm.

Abstract: Disclosed herein are an apparatus and method for adaptively accelerating a BLAS operation based on a GPU. The apparatus for adaptively accelerating a BLAS operation based on a GPU includes a BLAS operation acceleration unit for setting optimal OpenCL parameters using machine-learning data attribute information and OpenCL device information and for creating a kernel in a binary format by compiling kernel source code; an OpenCL execution unit for creating an OpenCL buffer for a BLAS operation using information about an OpenCL execution environment and the optimal OpenCL parameters and for accelerating machine learning in an embedded system in such a way that a GPU that is capable of accessing the created OpenCL buffer performs the BLAS operation using the kernel, and an accelerator application unit for returning the result of the BLAS operation to a machine-learning algorithm.

Abstract: Provided herein is a real-time QoS monitoring apparatus, including an application registration unit configured to register at least one monitoring target application program for QoS measurement; a function explorer unit configured to detect user-defined functions in application code of the at least one monitoring target application program; a loop-statement explorer unit configured to detect loop-statements in the application code; a user-defined location explorer unit configured to detect user-defined locations in the application code; and a heartbeat generator configured to generate a plurality of heartbeat calls to correspond to the functions detected by the function finder, the loop-statements detected by the loop finder, and the user-defined locations detected by the user-defined location finder. Accordingly, there are provided a real-time QoS monitoring apparatus and method, which may measure QoS in real time without additionally modifying the application program.

Abstract: Disclosed herein are an apparatus and method for controlling a governor based on a heterogeneous multicore system. The apparatus includes a heterogeneous core cluster unit for running any one of a first core cluster for high-performance operation and a second core cluster for low-power operation by switching therebetween; a governor-setting unit for generating operation setting information of a governor for controlling operation of the first core cluster and second core cluster; and a governor control unit for controlling operation of one or more governors based on the operation setting information.

Abstract: A method for configuring a monitoring environment of an application includes a monitoring location analysis step for detecting a monitoring location candidate, and a monitoring range analysis step for detecting a monitoring range, wherein the monitoring location analysis step includes receiving a source file of a general application, input m units of source lines, when an input source line is an execution line, calculating an execution load of the input source line, and determining a source line having the greatest execution load as a monitoring location candidate. Accordingly, it is possible to provide a method for determining an optimal performance monitoring location and an optimal performance monitoring range, which are required to configure an execution environment of a self-adaptive application, so as to perform efficient performance monitoring of the self-adaptive application.

Abstract: Provided herein is a real-time QoS monitoring apparatus, including an application registration unit configured to register at least one monitoring target application program for QoS measurement; a function explorer unit configured to detect user-defined functions in application code of the at least one monitoring target application program; a loop-statement explorer unit configured to detect loop-statements in the application code; a user-defined location explorer unit configured to detect user-defined locations in the application code; and a heartbeat generator configured to generate a plurality of heartbeat calls to correspond to the functions detected by the function finder, the loop-statements detected by the loop finder, and the user-defined locations detected by the user-defined location finder. Accordingly, there are provided a real-time QoS monitoring apparatus and method, which may measure QoS in real time without additionally modifying the application program.

Abstract: Disclosed are a method of providing an embedded software development tool for supporting an embedded software development environment based on native building, and an apparatus thereof. The method includes extracting information about a toolchain according to a request by the host system, and providing the extracted information to the host system, generating a project execution file by receiving a project, to which the information about the toolchain is applied, from the host system, and building the project, and generating a project execution result by executing the project execution file, thereby providing an embedded software development environment capable of minimizing the economic costs incurred for acquiring a cross toolchain corresponding to hardware and a platform of a target system and performing a development process when embedded software is developed.

Abstract: Provided are a tool for supporting open computing language (OpenCL) application software development for an embedded system and a supporting method thereof. The tool in conjunction with a target system includes a project management unit configured to separate and store OpenCL application software constituted of a host program and a kernel program into a host source code corresponding to the host program and a kernel source code corresponding to the kernel program, and manage a tool-chain corresponding to an OpenCL platform of the target system, and a builder configured to determine a compile type of the kernel source code and determine whether the kernel source code is built in response to the compile type. Accordingly, it is possible to rapidly provide a tool for supporting OpenCL application software development that can develop OpenCL application software for an embedded system.

Abstract: Disclosed are a method of providing an embedded software development tool for supporting an embedded software development environment based on native building, and an apparatus thereof. The method includes extracting information about a toolchain according to a request by the host system, and providing the extracted information to the host system, generating a project execution file by receiving a project, to which the information about the toolchain is applied, from the host system, and building the project, and generating a project execution result by executing the project execution file, thereby providing an embedded software development environment capable of minimizing the economic costs incurred for acquiring a cross toolchain corresponding to hardware and a platform of a target system and performing a development process when embedded software is developed.

Abstract: Disclosed herein is a method and apparatus for reconstructing an embedded software development environment. The apparatus includes an integrated management setting unit and an integrated project management unit. The integrated management setting unit sets at least one target supported by each of a plurality of development tools and registers the plurality of development tools. The integrated project management unit creates an integrated project in which the embedded software development environment is reconstructed using the plurality of development tools.

Abstract: Provided are a tool for supporting open computing language (OpenCL) application software development for an embedded system and a supporting method thereof. The tool in conjunction with a target system includes a project management unit configured to separate and store OpenCL application software constituted of a host program and a kernel program into a host source code corresponding to the host program and a kernel source code corresponding to the kernel program, and manage a tool-chain corresponding to an OpenCL platform of the target system, and a builder configured to determine a compile type of the kernel source code and determine whether the kernel source code is built in response to the compile type. Accordingly, it is possible to rapidly provide a tool for supporting OpenCL application software development that can develop OpenCL application software for an embedded system.

Abstract: Disclosed herein is a method and apparatus for reconstructing an embedded software development environment. The apparatus includes an integrated management setting unit and an integrated project management unit. The integrated management setting unit sets at least one target supported by each of a plurality of development tools and registers the plurality of development tools. The integrated project management unit creates an integrated project in which the embedded software development environment is reconstructed using the plurality of development tools.

Abstract: Provided is an integrated management apparatus and method for embedded software development tools, which can support all the development tools used in the process of developing embedded software under a single environment, and provide a developer with a development environment optimized to development work through the support of an integrated project and an integrated tool user interface (UI) on the basis of a development environment type. Thus, it is possible to increase efficiency of the embedded software development work.

Abstract: Provided are an apparatus and method for diagnosing operating system resources supporting universal serial bus (USB) device driver development for a Linux system. The apparatus and method device enable driver developers to simply perform, without professional knowledge of hardware and operating systems, operating system resource diagnosis such as detection and extraction of hardware information required for device driver development, search and change of various descriptor configurations, and process of data input/output through an endpoint. Thus, effort and cost required for a device driver development process are reduced, and it is possible to improve the efficiency of Linux USB device driver development work.

Abstract: Provided are a project management device and a method for an architecture modeling tool of AUTOSAR application software. The device includes an interface unit, a command execution unit, and a workspace management unit. The interface unit receives a processing command affecting project resources from a user, classifies the received command, and executes a corresponding call processing. The command execution unit analyzes a command inputted from the interface unit or delivered from a tool. If the command is a tool stop command, the device is stopped. If the command is an interface input command or is not the tool stop command, the command is classified into a description resource build command and a model update command to be executed. The workspace management unit creates, deletes, and updates contents of a workspace to reflect a processing result executed by the command execution unit on the workspace.

Abstract: A host system having a device driver development tool and a target system having a device driver agent, the tool and the target system separated from each other and interconnected with a communication network. The driver agent device communicates with the target system, receiving and processing various service requests from the device driver development tool, and transmits the processed services to the host system through the communication network. Thr driver agent determines types of the services requested from the device driver development tool and performs services based on the types of the services determined. Accordingly, the device driver can be adapted to the target system and tested without complex procedures to effectively reduce time and manpower needed for the device driver development.

Abstract: Provided are a device and method for executing a kernel application program interface (API) interactively in a Linux system, wherein a kernel API is independently operated in a user region to confirm an operation result. The device includes: a device interface unit for receiving a command from a user, and outputting an execution result corresponding to the command to the user; a kernel API command analysis unit for receiving the command from the device interface unit, analyzing the command, and generating kernel API execution information; and a kernel API command execution unit for executing a kernel API corresponding to the command using the kernel API execution information, and transmitting the execution result to the device interface unit. Thus, a kernel API can be applied using a simple process with reduced time and effort, instead of repeating the entire kernel program development process for a Linux system.