• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.238-241
• /
• 2005

최근 컴퓨터를 이용하여 효과적인 트레이드를 하려는 투자자들이 늘고 있다. 본 논문에서는 많은 인공지능 방법론 중에서 강화학습(reinforcement learning)을 이용하여 효과적으로 트레이딩하는 방법에 대해서 다루려한다. 특히 강화학습 중에서 natural policy gradient를 이용하여 actor의 파라미터를 업데이트하고, value function을 효과적으로 추정하기 위해 RLS(recursive least-squares) 기법으로 critic 부분을 업데이트하는 RLS 기반 natural actor-critic 알고리즘을 이용하여 트레이딩을 수행하는 전략에 대한 가능성을 살펴 보기로 한다.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.33 no.1
• /
• pp.19-33
• /
• 2008

In the previous double auction research for the market optimization, two basic assumptions are usually applied - (1) each trader has a linear or quasi-linear utility function of price and quantity, and (2) buyers as well as sellers have identical utility functions. However, in practice, each buyer and seller in a double auction market may have diverse utility functions for trading goods. Therefore, a flexible and integrated double auction mechanism that can integrate all traders' diverse utility functions is necessary. In particular, the flexible mechanism is more useful in a synchronous double auction because traders can properly change utilities in each round. Therefore, in this paper, we propose a flexible synchronous double auction mechanism in which traders can express diverse utility functions for the price and quantity of the goods, and optimal total market utility is guaranteed. In order to optimize the total market utility which consists of multiple complex utility functions of traders. We show the viability of the proposed mechanism through a several simulation experiments.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.82-84
• /
• 2016

We propose a new reliable SVD-based watermarking scheme having high fidelity and strong robustness with no false-positive problem. Each column of the principal component of a watermark image is embedded into singular values of LL, LH, HL and HH sub-bands of cover image with different scale factors. Each scale factor is optimized by trading-off fidelity and robustness using Differential Evolution (DE) algorithm. The proposed scheme improves fidelity and robustness of existing reliable SVD based watermarking schemes without any false-positive problem. Index Terms - watermarking, reliable SVD, DWT, principal component, Differential Evolution.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.9 no.4
• /
• pp.119-125
• /
• 2013

The amount of digital content grows exponentially by the development of the internet and the change of computing environments and the target also is getting wider. The industry using this digital content has been growing greatly. However, the distribution of pirated digital content is increasing using internet because digital content is easy to store and transmit and the damage is growing. In this paper, we propose safety trading system which can conceal the author's information safely in digital content in order to block illegal distribution of digital content. ARIA encryption algorithm is used to protect the concealed information of author in digital content and it is a help to track the illegal traders by doing fingerprinting of buyer information to digital content and managing the transaction information. The technical support for copyright dispute is to allow by providing the capability to verify illegal edit to original digital contents.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.87-89
• /
• 2018

We proposed an improved reliable SVD-based watermarking scheme resistant to geometric attacks while having high fidelity with no false-positive problem. Principal components of a watermark image are embedded into singular values of LL, LH, HL, and HH sub-bands of a transformed cover image by RDWT(redundant discrete wavelet transform) with optimal scale factors. Each scale factor is generated by trading-off fidelity and robustness using Differential Evolution (DE) algorithm. Zernike Moment (ZM) is used to estimate the geometric distortion and to correct the watermarked image before extracting watermark. The proposed scheme improves fidelity and robustness of existing reliable SVD based watermarking schemes while resisting to geometric attacks.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.599-600
• /
• 2023

본 연구는 한국 주식 시장인 KRX 데이터에 중점을 둔 투자 전략의 최적화를 목표로 하였다. 전통적인 기술 분석 방법은 투자자들의 경험에 의존하여 파라미터를 선택하였다. 하지만 이 연구에서는 기존의 경험에 기반한 파라미터 선택 대신 유전 알고리즘을 사용하여 파라미터를 최적화했다. 결과적으로, 이 전략은 상승장과 하락장 모두에서 buy-and-hold 전략보다 더 나은 성과를 보였다. 이는 기술 분석의 파라미터 최적화의 중요성을 강조하며, 더 효과적인 투자 전략 개발의 가능성을 보여준다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.9C
• /
• pp.1316-1322
• /
• 2004

Recently, the UWB (ultra wide-band) wireless communication technology, which provides high data transmission and is capable of linearly trading between throughput and signal-to-noise ratio (SNR), has drawn much attention for short-range wireless networks. Fully exploiting its notable features and minimizing its interference to coexisting other systems require the knowledge of SNR's at receivers In this paper, we propose an algorithm for estimating the pulse energy to noise ratio Ep/No of UWB signal with utilization of outputs from a correlator at a receiver, and evaluate the performance of the proposed algorithm through computer simulation. According to simulation results, the maximum standard deviation is about 1 13 dB with a block size of 500. Except for Ep/No=O and 2 dB cases with a block size of 500, no errors greater than 3 dB were observed in all the remaining experiments. Generally speaking, it improves as the true Ep/No, increases and as the block size increases A notable feature of the proposed algorithm is that it does not reduce the effective throughput because the estimation process does not require sending additional training signal of any specific format.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.39 no.2
• /
• pp.83-95
• /
• 2014

In this paper, we develop a portfolio selection model that can be used to invest in markets with margin requirements such as the foreign exchange market. An investment algorithm to implement the proposed portfolio selection model based on objective historical data is also presented. We further conduct empirical analysis on the performance of a hypothetical investment in the foreign exchange market, using the proposed portfolio selection model and investment algorithm. Using 7 currency pairs that recorded the highest trading volume in the foreign exchange market during the most recent 10 years, we compare the performance of 1) the Dollar Index, 2) a 1/N Portfolio which equally allocates capital to all N assets considered for investment, and 3) a hypothetical investment portfolio selected and managed according to the portfolio selection model and investment algorithm proposed in this paper. Performance is compared in terms of accumulated returns and Sharpe ratios for the 10-year period from January 2003 to December 2012. The results show that the hypothetical investment portfolio outperforms both benchmarks, with superior performance especially during the period following financial crisis. Overall, this paper suggests that a mathematical approach for selecting and managing an optimal investment portfolio based on objective data can achieve outstanding performance in the foreign exchange market.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.3
• /
• pp.477-489
• /
• 2014

Based on some flaws occurred for implementing a public key cryptosystem in the embedded security device, many side-channel attacks to extract the secret private key have been tried. In spite of the fact that the cryptographic exponentiation is basically composed of a sequence of multiplications and squarings, a new Square Always exponentiation algorithm was recently presented as a countermeasure against side-channel attacks based on trading multiplications for squarings. In this paper, we propose Known Power Collision Analysis and modified Doubling attacks to break the Right-to-Left Square Always exponentiation algorithm which is known resistant to the existing side-channel attacks. And we also present a Collision-based Combined Attack which is a combinational method of fault attack and power collision analysis. Furthermore, we verify that the Square Always algorithm is vulnerable to the proposed side-channel attacks using computer simulation.

• Journal of Intelligence and Information Systems
• /
• v.20 no.4
• /
• pp.43-58
• /
• 2014

As the use of trading systems has increased rapidly, many researchers have become interested in developing effective stock market prediction models using artificial intelligence techniques. Stock market prediction involves multifaceted interactions between market-controlling factors and unknown random processes. A successful stock prediction model achieves the most accurate result from minimum input data with the least complex model. In this research, we develop a combination model of ${\pi}$-fuzzy logic and support vector machine (SVM) models, using a genetic algorithm to optimize the parameters of the SVM and ${\pi}$-fuzzy functions, as well as feature subset selection to improve the performance of stock market prediction. To evaluate the performance of our proposed model, we compare the performance of our model to other comparative models, including the logistic regression, multiple discriminant analysis, classification and regression tree, artificial neural network, SVM, and fuzzy SVM models, with the same data. The results show that our model outperforms all other comparative models in prediction accuracy as well as return on investment.