Automatic Configuration Research Articles

MOODY: An ontology-driven framework for standardizing multi-objective evolutionary algorithms

The application of semantic technologies, particularly ontologies, in the realm of multi-objective evolutionary algorithms is overlook despite their effectiveness in knowledge representation. In this paper, we introduce MOODY, an ontology specifically tailored to formalize these kinds of algorithms, encompassing their respective parameters, and multi-objective optimization problems based on a characterization of their search space landscapes. MOODY is designed to be particularly applicable in automatic algorithm configuration, which involves the search of the parameters of an optimization algorithm to optimize its performance. In this context, we observe a notable absence of standardized components, parameters, and related considerations, such as problem characteristics and algorithm configurations. This lack of standardization introduces difficulties in the selection of valid component combinations and in the re-use of algorithmic configurations between different algorithm implementations. MOODY offers a means to infuse semantic annotations into the configurations found by automatic tools, enabling efficient querying of the results and seamless integration across diverse sources through their incorporation into a knowledge graph. We validate our proposal by presenting four case studies.

Intra- and Inter-expert Validation of an Automatic Segmentation Method for Fluid Regions Associated with Central Serous Chorioretinopathy in OCT Images

Central Serous Chorioretinopathy (CSC) is a retinal disorder caused by the accumulation of fluid, resulting in vision distortion. The diagnosis of this disease is typically performed through Optical Coherence Tomography (OCT) imaging, which displays any fluid buildup between the retinal layers. Currently, these fluid regions are manually detected by visual inspection a time-consuming and subjective process that can be prone to errors. A series of six deep learning-based automatic segmentation architectural configurations of different levels of complexity were trained and compared in order to determine the best model intended for the automatic segmentation of CSC-related lesions in OCT images. The best performing models were then evaluated in an external validation study. Furthermore, an intra- and inter-expert analysis was conducted in order to compare the manual segmentation performed by expert ophthalmologists with the automatic segmentation provided by the models. Test results of the best performing configuration achieved a mean Dice of 0.868±0.056\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$0.868 \\mathop {\\pm } 0.056$$\\end{document} in the internal dataset. In the external validation set, these models achieved a level of agreement with human experts of up to 0.960 in terms of Kappa coefficient, contrasting with a value of 0.951 for agreement between human experts. Overall, the models reached a better agreement with either of the human experts than these experts with each other, suggesting that automatic segmentation models for the detection of CSC-related lesions in OCT imaging can be useful tools for assessing this disease, reducing the workload of manual inspection and leading to a more robust and objective diagnosis method.

EPyDGGA: automatic configuration for fitting epidemic curves

Many epidemiological models and algorithms are used to fit the parameters of a given epidemic curve. On many occasions, fitting algorithms are interleaved with the actual epidemic models, which yields combinations of model-parameters that are hard to compare among themselves. Here, we provide a model-agnostic framework for epidemic parameter fitting that can (fairly) compare different epidemic models without jeopardizing the quality of the fitted parameters. Briefly, we have developed a Python framework that expects a Python function (epidemic model) and epidemic data and performs parameter fitting using automatic configuration. Our framework is capable of fitting parameters for any type of epidemic model, as long as it is provided as a Python function (or even in a different programming language). Moreover, we provide the code for different types of models, as well as the implementation of 4 concrete models with data to fit them. Documentation, code and examples can be found at https://ulog.udl.cat/static/doc/epidemic-gga/html/index.html.

Study and evaluation of automatic offloading for function blocks of applications

Systems using graphical processing units (GPUs) and field-programmable gate arrays (FPGAs) have increased due to their advantages over central processing units (CPUs). However, such systems require the understanding of hardware-specific technical specifications such as Hardware Description Language (HDL) and compute unified device architecture (CUDA), which is a high hurdle. Based on this background, we previously proposed environment-adaptive software that enables automatic conversion, configuration and high-performance operation of existing code according to the hardware to be placed. As an element of this concept, we also proposed a method of automatically offloading loop statements of application source code for CPUs to GPUs and FPGAs. In this paper, we propose a method for offloading a function block, which is a larger unit, instead of individual loop statements in an application to achieve higher speed by automatically offloading to GPUs and FPGAs. We implemented the proposed method and evaluated it using current applications offloading to GPUs and FPGAs.

AutoPEFT: Automatic Configuration Search for Parameter-Efficient Fine-Tuning

<scp>AutoPEFT</scp>: Automatic Configuration Search for Parameter-Efficient Fine-Tuning

Managing software build infrastructure at ALICE using Hashicorp Nomad

Managing software build infrastructure at ALICE using Hashicorp Nomad

A STUDY OF RAPID MAPPING TECHNOLOGY BASED ON ADOBE ILLUSTRATOR

Abstract. At present, China's urban and rural construction, national security, emergency and disaster relief have put forward higher and faster requirements for high-quality and current maps. How to use the latest spatial database in the shortest possible time to produce and provide high-quality, highly presentable, content-rich maps is an urgent issue. In this paper, we study the rapid mapping technology based on GIS platform and map mapping software, study the automatic map mapping method under AI environment, design and develop the spatial basic geographic database-driven rapid mapping system, realize the automatic lossless conversion of GIS data to map mapping software according to the established mapping rule base, realize the automatic configuration of symbols and annotations, and realize the automation of processing of element relationship. The AI-based database-driven rapid mapping technology has been used in the practice of map compilation in prefecture-level cities, replacing most of the manual repetitive operations, greatly improving the efficiency of map compilation, and enhancing the capacity of emergency mapping services.

Automatic configuration analysis method of planetary gear automatic transmission

Planetary gear trains (PGT) with the aim of achieving multiple speed ratios are widely used in automatic transmission (AT). A fast configuration analysis method is of significance for improving the efficiency of AT design, and it is essential to study the automatic configuration analysis method for AT. For this purpose, the AT configuration is represented by a matrix model, which includes the structural matrix of PGT and shifting elements (SEs) matrix. Based on the proposed matrix representation, the speed relation and torque relation matrices of AT are automatically derived to analyze the kinematics and dynamics. Five performance indexes of AT configuration, specifically transmission ratio, rotation speed of components, torque transmitted by SEs, power flow, and transmission efficiency, are automatically calculated through the analysis result. The entire analysis process can be completed in 1 s on the computer. The proposed matrix model offers a unified representation for the configuration of AT derived from different synthesis algorithms. It facilitates rapid automated configuration analysis and performance evaluation, thereby positively influencing the design of automatic transmissions. Furthermore, the matrix model can be applied to the configuration synthesis of AT.

On the automatic design of multi-objective particle swarm optimizers: experimentation and analysis

Research in multi-objective particle swarm optimizers (MOPSOs) progresses by proposing one new MOPSO at a time. In spite of the commonalities among different MOPSOs, it is often unclear which algorithmic components are crucial for explaining the performance of a particular MOPSO design. Moreover, it is expected that different designs may perform best on different problem families and identifying a best overall MOPSO is a challenging task. We tackle this challenge here by: (1) proposing AutoMOPSO, a flexible algorithmic template for designing MOPSOs with a design space that can instantiate thousands of potential MOPSOs; and (2) searching for good-performing MOPSO designs given a family of training problems by means of an automatic configuration tool (irace). We apply this automatic design methodology to generate a MOPSO that significantly outperforms two state-of-the-art MOPSOs on four well-known bi-objective problem families. We also identify the key design choices and parameters of the winning MOPSO by means of ablation. AutoMOPSO is publicly available as part of the jMetal framework.

Taming complexity of industrial printing systems using a constraint‐based DSL: An industrial experience report

AbstractFlexible printing systems are highly complex systems that consist of printers, that print individual sheets of paper, and finishing equipment, that processes sheets after printing, for example, assembling a book. Integrating finishing equipment with printers involves the development of control software that configures the devices, taking hardware constraints into account. This control software is highly complex to realize due to (1) the intertwined nature of printing and finishing, (2) the large variety of print products and production options for a given product, and (3) the large range of finishers produced by different vendors. We have developed a domain‐specific language called CSX that offers an interface to constraint solving specific to the printing domain. We use it to model printing and finishing devices and to automatically derive constraint solver‐based environments for automatic configuration. We evaluate CSX on its coverage of the printing domain in an industrial context, and we report on lessons learned on using a constraint‐based DSL in an industrial context.

Automatic MILP solver configuration by learning problem similarities

Automatic MILP solver configuration by learning problem similarities

Model-driven engineering for low-code ground support equipment configuration and automatic test procedures definition

Space domain systems must go through different types of ground testing. For system-level black-box functional testing, ground support equipment is built ad hoc, customized for each different mission. Building any of this specific equipment involves considerable engineering effort that can be diluted using techniques that allow reusing configurations of the test setup environment without recodification. These techniques can be merged into a hierarchical checking process based on independent filters to compare the received outputs with the expected ones. The hierarchical checking allows the separate definition of several layers or levels of validation, from the lowest protocol packet levels to the logical abstractions at the highest application level. This paper introduces a solution based on Model-Driven Engineering for Low-Code Ground Support Equipment. It uses the abovementioned mechanisms to reduce the effort to develop and customize ground support systems for different missions. This solution is integrated within an environment called MASSIVA, which allows the automatic configuration and definition of system-level test procedures. This environment has been used in the software verification and validation process of the Instrument Control Unit of the Energetic Particle Detector on board Solar Orbiter.

AutoInit: Analytic Signal-Preserving Weight Initialization for Neural Networks

Neural networks require careful weight initialization to prevent signals from exploding or vanishing. Existing initialization schemes solve this problem in specific cases by assuming that the network has a certain activation function or topology. It is difficult to derive such weight initialization strategies, and modern architectures therefore often use these same initialization schemes even though their assumptions do not hold. This paper introduces AutoInit, a weight initialization algorithm that automatically adapts to different neural network architectures. By analytically tracking the mean and variance of signals as they propagate through the network, AutoInit appropriately scales the weights at each layer to avoid exploding or vanishing signals. Experiments demonstrate that AutoInit improves performance of convolutional, residual, and transformer networks across a range of activation function, dropout, weight decay, learning rate, and normalizer settings, and does so more reliably than data-dependent initialization methods. This flexibility allows AutoInit to initialize models for everything from small tabular tasks to large datasets such as ImageNet. Such generality turns out particularly useful in neural architecture search and in activation function discovery. In these settings, AutoInit initializes each candidate appropriately, making performance evaluations more accurate. AutoInit thus serves as an automatic configuration tool that makes design of new neural network architectures more robust. The AutoInit package provides a wrapper around TensorFlow models and is available at https://github.com/cognizant-ai-labs/autoinit.

Towards Green Automated Machine Learning: Status Quo and Future Directions

Automated machine learning (AutoML) strives for the automatic configuration of machine learning algorithms and their composition into an overall (software) solution — a machine learning pipeline — tailored to the learning task (dataset) at hand. Over the last decade, AutoML has developed into an independent research field with hundreds of contributions. At the same time, AutoML is being criticized for its high resource consumption as many approaches rely on the (costly) evaluation of many machine learning pipelines, as well as the expensive large-scale experiments across many datasets and approaches. In the spirit of recent work on Green AI, this paper proposes Green AutoML, a paradigm to make the whole AutoML process more environmentally friendly. Therefore, we first elaborate on how to quantify the environmental footprint of an AutoML tool. Afterward, different strategies on how to design and benchmark an AutoML tool w.r.t. their “greenness”, i.e., sustainability, are summarized. Finally, we elaborate on how to be transparent about the environmental footprint and what kind of research incentives could direct the community in a more sustainable AutoML research direction. As part of this, we propose a sustainability checklist to be attached to every AutoML paper featuring all core aspects of Green AutoML.

Digital Substations and IEC 61850: A Primer

Modern electrical substations are managed by intelligent electronic devices communicating through Ethernet-based networks. Because of the highly critical roles of substations, which include safety and protection, electronic devices need to be configured automatically through highly reliable and low-latency networks. This gives substation networks some requirements not addressed by previous industrial network standards. To address these unique challenges, IEC 61850 provides a collection of protocols and a model for the automatic configuration of digital substations. This article briefly overviews this standard and the security efforts to protect substations from attacks.

XFilter: An Extension of the Integrity Measurement Architecture Based on Fine-Grained Policies

The Integrity Measurement Architecture subsystem on the Linux platform is a critical security component in the kernel to ensure the integrity of the running system. However, the default Integrity Measurement Architecture policy mechanisms based on options such as file owner and FSMAGIC cannot achieve a file-level configuration. Although Integrity Measurement Architecture supports the Linux Security Module policy rules to be close to the goal of fine-grained configuration, it is not easy to be managed because the Linux Security Module was not originally designed for integrity measurement. Moreover, the Linux Security Module-based policy does not apply in some use cases considering the type of Mandatory Access Control tools chosen by users. This paper presents a new policy configuration option, named XFilter, that achieves a fine-grained policy configuration method. The XFilter includes two policy matching mechanisms, XLabel and XList, which share the same policy token created for XFilter exclusively. XLabel marks the files for measurement using a label in the file’s extended attribute (xattr). By contrast, XList stores the measurement information in a list of file paths. To simplify the deployment, an automatic configuration process is implemented for integrating into the package management system. The evaluation results suggest that both mechanisms satisfy the requirements of file-level IMA policy control and create a performance burden for system operation in the acceptable range. They also reveal a positive correlation between the increment of the system latency and the growth of the length of file paths list for the XList mechanism.

Construction technology of graph and model mutual capacity in smart substation

In order to meet the requirements of automatic mapping of the smart substation models, based on the application requirements of the smart substation model construction and automatic model association configuration, the automatic generation technology of smart station diagram models is studied, and the automatic correlation modeling method and wiring diagram generation technology of primary and secondary equipment models based on the automatic matching of predefined correlation models are proposed. Support the realization of the goal of effective configuration and efficient application of the secondary system model of the smart substation.

A Novel OpenBCI Framework for EEG-Based Neurophysiological Experiments.

An Open Brain-Computer Interface (OpenBCI) provides unparalleled freedom and flexibility through open-source hardware and firmware at a low-cost implementation. It exploits robust hardware platforms and powerful software development kits to create customized drivers with advanced capabilities. Still, several restrictions may significantly reduce the performance of OpenBCI. These limitations include the need for more effective communication between computers and peripheral devices and more flexibility for fast settings under specific protocols for neurophysiological data. This paper describes a flexible and scalable OpenBCI framework for electroencephalographic (EEG) data experiments using the Cyton acquisition board with updated drivers to maximize the hardware benefits of ADS1299 platforms. The framework handles distributed computing tasks and supports multiple sampling rates, communication protocols, free electrode placement, and single marker synchronization. As a result, the OpenBCI system delivers real-time feedback and controlled execution of EEG-based clinical protocols for implementing the steps of neural recording, decoding, stimulation, and real-time analysis. In addition, the system incorporates automatic background configuration and user-friendly widgets for stimuli delivery. Motor imagery tests the closed-loop BCI designed to enable real-time streaming within the required latency and jitter ranges. Therefore, the presented framework offers a promising solution for tailored neurophysiological data processing.

Fast Access Method of IoT Smart Terminal Based On C/S Architecture

In order to achieve efficient access and fast access of intelligent terminals, this paper proposes a fast access method for intelligent terminals of the Internet of Things based on C/S architecture. First, a terminal auto configuration program is developed; The automatic configuration program sends application configuration information to the configuration server; Then, the server verifies the identity of the terminal and feeds back the corresponding configuration information. Finally, the automatic configuration program automatically configures the terminal according to the feedback information and accesses the Internet of Things management platform. The experimental results show that this method can realize remote distribution of terminal configuration information and automatic configuration of terminals. Compared with traditional methods, the terminal access time can be reduced by about 70%.

Automatic Configuration of an Order as an Integral Part of a Cyber-Physical System in a Manufacturing Operating According to Mass-Customisation Strategy

The contemporary consumer market determines the use of mass customisation by manufacturers. Therefore, from the point of view of Industry 4.0 (I4.0), concept data and communications and analytics are relevant to the implementation of the mass-customisation strategy. The problem involves at least three subjects: how to connect the currently available information system within an enterprise with new I4.0 technologies, how new information solutions can support the verification of the feasibility of tailoring an order to the needs of the customer, and, finally, how to model a knowledge base for a cyber-physical system containing a formal record of the knowledge acquired regarding customer preferences. Therefore, in this paper, we developed a new algorithm that will enable, on one hand, the collection and recording of customer preferences, and, on the other hand, the integration of those data and information that are available within the Enterprise Resource Planning (ERP) system. The main contribution of this work is the use of specialist knowledge and data from ERP regarding production capabilities implemented in a manufacturing enterprise in order to model the scenario of generating possible orders for a client, and finally applying the new solution for the operation of manufacturing according to mass-customisation strategy in a real company that manufactures mattresses.