• Electronics and Telecommunications Trends
• /
• v.36 no.2
• /
• pp.32-42
• /
• 2021

The rapid growth of deep-learning applications has invoked the R&D of artificial intelligence (AI) processors. A dedicated software framework such as a compiler and runtime APIs is required to achieve maximum processor performance. There are various compilers and frameworks for AI training and inference. In this study, we present the features and characteristics of AI compilers, training frameworks, and inference engines. In addition, we focus on the internals of compiler frameworks, which are based on either basic linear algebra subprograms or intermediate representation. For an in-depth insight, we present the compiler infrastructure, internal components, and operation flow of ETRI's "AI-Ware." The software framework's significant role is evidenced from the optimized neural processing unit code produced by the compiler after various optimization passes, such as scheduling, architecture-considering optimization, schedule selection, and power optimization. We conclude the study with thoughts about the future of state-of-the-art AI compilers.

• Microbiology and Biotechnology Letters
• /
• v.51 no.1
• /
• pp.69-82
• /
• 2023

Nitrilases are a hydrolase group of enzymes that catalyzes nitrile compounds and produce industrially important organic acids. The current objective is to optimize nitrilase production using statistical methods assisted with artificial intelligence (AI) tool from novel nitrile degrading isolate. A nitrile hydrolyzing bacteria Bacillus subtilis AGAB-2 (GenBank Ascension number- MW857547) was isolated from industrial effluent waste through an enrichment culture technique. The culture conditions were optimized by creating an orthogonal design with 7 variables to investigate the effect of the significant factors on nitrilase activity. On the basis of obtained data, an AI-driven support vector machine was used for the fitted regression, which yielded new sets of predicted responses with zero mean error and reduced root mean square error. The results of the above global optimization were regarded as the theoretical optimal function conditions. Nitrilase activity of 9832 ± 15.3 U/ml was obtained under optimized conditions, which is a 5.3-fold increase in compared to unoptimized (1822 ± 18.42 U/ml). The statistical optimization method involving Plackett Burman Design and Response surface methodology in combination with an AI tool created a better response prediction model with a significant improvement in enzyme production.

• Corrosion Science and Technology
• /
• v.23 no.2
• /
• pp.131-138
• /
• 2024

This paper introduces Primetals Technologies' Through-Process Optimization (TPO) Services and Through-Process Quality Control (TPQC) System, which integrate domain knowledge, software, and automation expertise to assist steel producers in achieving operational excellence. TPQC collects high-resolution process and product data from the entire production route, providing visualizations and facilitating quality assurance. It also enables the application of artificial intelligence techniques to optimize processes, accelerate steel grade development, and enhance product quality. The main objective of TPO is to grow and digitize operational know-how, increase profitability, and better meet customer needs. The paper describes the contribution of these systems to achieving operational excellence, with a focus on quality assurance. Transparent and traceable production data is used for manual and automatic quality evaluation, resulting in product quality status and guiding the product disposition process. Deviation management is supported by rule-based and AI-based assistants, along with monitoring, alarming, and reporting functions ensuring early recognition of deviations. Embedded root cause proposals and their corrective and compensatory actions facilitate decision support to maintain product quality. Quality indicators and predictive quality models further enhance the efficiency of the quality assurance process. Utilizing the quality assurance software package, TPQC acts as a "one-truth" platform for product quality key players.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.3
• /
• pp.85-92
• /
• 2023

Recently, interest in artificial intelligence has increased due to the development of artificial intelligence fields such as ChatGPT and self-driving cars. However, there are still many unknown elements in training process of artificial intelligence, so that optimizing the model requires more time and effort than it needs. Therefore, there is a need for a tool or methodology that can analyze the weight changes during the training process of artificial intelligence and help out understatnding those changes. In this research, I propose a visualization system which helps people to understand the accumulated weight changes. The system calculates the weights for each training period to accumulates weight changes and stores accumulated weight changes to plot them in 3D space. This research will allow us to explore different aspect of artificial intelligence learning process, such as understanding how the model get trained and providing us an indicator on which hyperparameters should be changed for better performance. These attempts are expected to explore better in artificial intelligence learning process that is still considered as unknown and contribute to the development and application of artificial intelligence models.

• Journal of KIISE
• /
• v.41 no.12
• /
• pp.1090-1098
• /
• 2014

In this paper, an improved ACDE (Adaptive Cauchy Differential Evolution) algorithm with faster convergence speed, called ACDE2, is suggested. The baseline ACDE algorithm uses a "DE/rand/1" mutation strategy to provide good population diversity, and it is appropriate for solving multimodal optimization problems. However, the convergence speed of the mutation strategy is slow, and it is therefore not suitable for solving unimodal optimization problems. The ACDE2 algorithm uses a "DE/current-to-best/1" mutation strategy in order to provide a fast convergence speed, where a control parameter initialization operator is used to avoid converging to local optimization. The operator is executed after every predefined number of generations or when every individual fails to evolve, which assigns a value with a high level of exploration property to the control parameter of each individual, providing additional population diversity. Our experimental results show that the ACDE2 algorithm performs better than some state-of-the-art DE algorithms, particularly in unimodal optimization problems.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.2
• /
• pp.236-244
• /
• 2005

In this study, we introduce a new topology of Fuzzy Polynomial Neural Networks(FPNN) that is based on fuzzy relation and evolutionally optimized Multi-Layer Perceptron, discuss a comprehensive design methodology and carry out a series of numeric experiments. The construction of the evolutionally optimized FPNN(EFPNN) exploits fundamental technologies of Computational Intelligence. The architecture of the resulting EFPNN results from a synergistic usage of the genetic optimization-driven hybrid system generated by combining rule-based Fuzzy Neural Networks(FNN) with polynomial neural networks(PNN). FNN contributes to the formation of the premise part of the overall rule-based structure of the EFPNN. The consequence part of the EFPNN is designed using PNN. As the consequence part of the EFPNN, the development of the genetically optimized PNN(gPNN) dwells on two general optimization mechanism: the structural optimization is realized via GAs whereas in case of the parametric optimization we proceed with a standard least square method-based learning. To evaluate the performance of the EFPNN, the models are experimented with the use of several representative numerical examples. A comparative analysis shows that the proposed EFPNN are models with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• Communications of Mathematical Education
• /
• v.37 no.1
• /
• pp.65-84
• /
• 2023

The method of Lagrange multipliers, one of the most fundamental algorithms for solving equality constrained optimization problems, has been widely used in basic mathematics for artificial intelligence (AI), linear algebra, optimization theory, and control theory. This method is an important tool that connects calculus and linear algebra. It is actively used in artificial intelligence algorithms including principal component analysis (PCA). Therefore, it is desired that instructors motivate students who first encounter this method in college calculus. In this paper, we provide an integrated perspective for instructors to teach the method of Lagrange multipliers effectively. First, we provide visualization materials and Python-based code, helping to understand the principle of this method. Second, we give a full explanation on the relation between Lagrange multiplier and eigenvalues of a matrix. Third, we give the proof of the first-order optimality condition, which is a fundamental of the method of Lagrange multipliers, and briefly introduce the generalized version of it in optimization. Finally, we give an example of PCA analysis on a real data. These materials can be utilized in class for teaching of the method of Lagrange multipliers.

• Journal of Information Technology Services
• /
• v.19 no.2
• /
• pp.109-124
• /
• 2020

Facing the 4^th Industrial Revolution era, researches on artificial intelligence have become active and attempts have been made to apply machine learning in various fields. In the field of finance, Robo Advisor service, which analyze the market, make investment decisions and allocate assets instead of people, are rapidly expanding. The stock price prediction using the machine learning that has been carried out to date is mainly based on the prediction of the market index such as KOSPI, and utilizes technical data that is fundamental index or price derivative index using financial statement. However, most researches have proceeded without any explicit verification of the prediction rate of the learning data. In this study, we conducted an experiment to determine the degree of market prediction ability of basic indicators, technical indicators, and system risk indicators (AR) used in stock price prediction. First, we set the core parameters for each financial indicator and define the objective function reflecting the return and volatility. Then, an experiment was performed to extract the sample from the distribution of each parameter by the Markov chain Monte Carlo (MCMC) method and to find the optimum value to maximize the objective function. Since Robo Advisor is a commodity that trades financial instruments such as stocks and funds, it can not be utilized only by forecasting the market index. The sample for this experiment is data of 17 years of 1,500 stocks that have been listed in Korea for more than 5 years after listing. As a result of the experiment, it was possible to establish a meaningful trading strategy that exceeds the market return. This study can be utilized as a basis for the development of Robo Advisor products in that it includes a large proportion of listed stocks in Korea, rather than an experiment on a single index, and verifies market predictability of various financial indicators.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.12
• /
• pp.5904-5927
• /
• 2019

Many artificial intelligence (AI) techniques have been embedded into various engineering technologies to assist them in achieving different goals. The integration of modern technologies with energy consumption management system and occupant's comfort inside buildings results in the introduction of intelligent building concept. The major aim of this integration is to manage the energy consumption effectively and keeping the occupant satisfied with the internal environment of the building. The last few couple of years have seen many applications of AI techniques for optimizing the energy consumption with maximizing the user comfort in smart buildings but still there is much room for improvement in this area. In this paper, a hybrid of two AI algorithms called firefly algorithm (FA) and genetic algorithm (GA) has been used for user comfort maximization with minimum energy consumption inside smart building. A complete user friendly system with data from various sensors, user, processes, power control system and different actuators is developed in this work for reducing power consumption and increase the user comfort. The inputs of optimization algorithms are illumination, temperature and air quality sensors' data and the user set parameters whereas the outputs of the optimization algorithms are optimized parameters. These optimized parameters are the inputs of different fuzzy controllers which change the status of different actuators according to user satisfaction.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.529-533
• /
• 2019

Combined cycle power plants are often used to produce power. These days prediction of power plant output based on operating parameters is a major concern. This paper presents an approach to using computational intelligence technique to predict the output power of combined cycle power plant. Computational intelligence techniques have been developed and applied to many real world problems. In this paper, tree architectures of fuzzy neural networks are considered to predict the output power. Tree architectures of fuzzy neural networks have an advantage of reducing the number of rules by selecting fuzzy neurons as nodes and relevant inputs as leaves optimally. For the optimization of the networks, two-step optimization method is used. Genetic algorithms optimize the binary structure of the networks by selecting the nodes and leaves as binary, and followed by random signal-based learning further refines the optimized binary connections in the unit interval. To verify the effectiveness of the proposed method, combined cycle power plant dataset obtained from the UCI Machine Learning Repository Database is considered.