• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.150-152
• /
• 2022

In this work, we propose a model that considers the behavior and synaptic plasticity of sensory neurons based on Liquid Time-constant Network (LTC). The neuron connection structure was experimented with four types: the increasing number of neurons, the decreasing number, the decreasing number, and the decreasing number. In this study, we experimented using a time series prediction dataset to see if the performance of the changed model improved compared to LTC. Experimental results show that the application of modeling of sensory neurons does not always bring about performance improvements, but improves performance through proper selection of learning rules depending on the type of dataset. In addition, the connective structure of neurons showed improved performance when it was less than four layers.

• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.1002-1012
• /
• 2019

360VR content requires short latency, such as immediate response to viewers' viewport changes and high quality video delivery. It is necessary to consider efficient transmission that guarantees the QoE(Quality of Experience) of the 360VR contents with limited bandwidth. Several research has been introduced to reduce overall bandwidth consumption by predicting a user's viewport and allocating different bit rates to the area corresponding to the viewport. In this paper, we propose novel viewport prediction scheme that uses sound source location information of 360VR contents as auditory recognition information along with visual recognition information. Also, we propose efficient transmission algorithm by allocating a bit rate properly based on improved viewport prediction. The proposed scheme improves the accuracy of the viewport prediction and provides high quality videos to tiles corresponding to the user's viewpoint within the limited bandwidth.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.5
• /
• pp.48-53
• /
• 2001

In speaker verification based on LPC analysis the prediction residues are ignored and LPCC(LPC cepstrum) are only used to compose feature vectors. In this study, LPCC and RCEP (residual cepstrum) extracted from residues are used as feature parameters in the various environmental speaker verification. We propose the weighting function which can enlarge inter-speaker variation by weighting pitch, speaker inherent vector, included in residual cepstrum. Simulation results show that the average speaker verification rate is improved in the rate of 6％ with RCEP and LPCC at the same time and is improved in the rate of 2.45% with the proposed weighted RCEP and LPCC at the same time compared with no weighting.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2009.05a
• /
• pp.1265-1271
• /
• 2009

본 연구에서는 강우의 시 공간적 분포의 불규칙한 변동성을 고려한 강우-유출예측모형을 위해 인공신경망(Artificial Neural Networks: ANNs)의 기법의 일종인 자기조직화(Self Organizing Map: SOM) 이론과 역전파 학습 알고리즘(Back Propagation Algorithm: BPA) 이론을 복합적으로 이용하였다. 기존의 인공신경망 연구에서 야기된 저 갈수기의 유출량에 대한 과대평가, 홍수기의 유출량에 대한 과소평가, 예측값이 연속적으로 선행 유출량을 나타내는 Persistence 현상을 해결하기 위하여 패턴분류 성능을 지닌 SOM 이론을 예측모형의 전처리 과정으로 이용하였다. 먼저, 본 연구에서 제안한 방법은 SOM에 의해 강우-유출 관계를 분류하고, SOM에 의한 분류에 따라 각각의 모형을 구성한다. 개별적으로 구축된 모형은 유출량의 예측을 위해 각각의 양상에 따라 분류된 자료를 이용한다. 결과적으로 본 연구에서 제안한 방법은 과거의 인공신경망의 일반적인 적용에 의한 결과보다 더 나은 예측능력을 보여주었으며, 더불어 유출량의 과소 및 과대추정과 Persistence 현상과 같은 문제점이 나타나지 않았다. 또한 강우량 및 유출량의 범위에 제한을 받지 않는 강우-유출예측 모형의 개발 및 홍수기로부터 갈수기까지의 보다 넓은 범위의 유출량의 예측에 기여할 것으로 기대된다.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.12
• /
• pp.485-496
• /
• 2018

So far, the deep learning, a field of artificial intelligence, has achieved remarkable results in solving problems from unstructured data. However, it is difficult to comprehensively judge situations like humans, and did not reach the level of intelligence that deduced their relations and predicted the next situation. Recently, deep neural networks show that artificial intelligence can possess powerful relational reasoning that is core intellectual ability of human being. In this paper, to analyze and observe the performance of Relation Networks (RN) among the neural networks for relational reasoning, two types of RN-based deep neural network models were constructed and compared with the baseline model. One is a visual question answering RN model using Sort-of-CLEVR and the other is a text-based question answering RN model using bAbI task. In order to maximize the performance of the RN-based model, various performance improvement experiments such as hyper parameters tuning have been proposed and performed. The effectiveness of the proposed performance improvement methods has been verified by applying to the visual QA RN model and the text-based QA RN model, and the new domain model using the dialogue-based LL dataset. As a result of the various experiments, it is found that the initial learning rate is a key factor in determining the performance of the model in both types of RN models. We have observed that the optimal initial learning rate setting found by the proposed random search method can improve the performance of the model up to 99.8%.

• Journal of Digital Convergence
• /
• v.16 no.9
• /
• pp.147-157
• /
• 2018

Since Google launched a prediction service for influenza-like illness(ILI), studies on ILI prediction based on web search data have proliferated worldwide. In this regard, this study aims to build short-term predictive models for ILI in Korea using ILI and web search data and measure the performance of the said models. In these proposed ILI predictive models specific to Korea, ILI surveillance data of Korea CDC and Korean web search data of Google and Naver were used along with the ARIMA model. Model 1 used only ILI data. Models 2 and 3 added Google and Naver search data to the data of Model 1, respectively. Model 4 included a common query used in Models 2 and 3 in addition to the data used in Model 1. In the training period, the goodness of fit of all predictive models was higher than 95% ($R^2$). In predictive periods 1 and 2, Model 1 yielded the best predictions (99.98% and 96.94%, respectively). Models 3(a), 4(b), and 4(c) achieved stable predictability higher than 90% in all predictive periods, but their performances were not better than that of Model 1. The proposed models that yielded accurate and stable predictions can be applied to early warning systems for the influenza pandemic in Korea, with supplementary studies on improving their performance.

• Journal of Internet Computing and Services
• /
• v.14 no.5
• /
• pp.59-67
• /
• 2013

In this paper, we propose the LAR_CF, latent attribute rating-based collaborative filtering, that is robust to data sparsity problem which is one of traditional problems caused of decreasing rating prediction accuracy. As compared with that existing collaborative filtering method uses a preference rating rated by users as feature vector to calculate similarity between objects, the proposed method improves data sparsity problem using unique attributes of two target objects with existing explicit preference. We consider MovieLens 100k dataset and its item attributes to evaluate the LAR_CF. As a result of artificial data sparsity and full-rating experiments, we confirmed that rating prediction accuracy can be improved rating prediction accuracy in data sparsity condition by the LAR_CF.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.22-22
• /
• 2015

가뭄은 적시에 경보해야 하는 홍수와 달리 진행속도가 느리고 시간적으로 대처할 여유가 있어 진행중일지라도 미리 감지만 한다면 그 피해를 최소화할 수 있다. 이로 인해 미국 등 수문기상 선진국에서는 수문기상 장기예보자료로부터 가뭄전망정보 생산기술을 개발하였으며, 특히 가뭄전망의 정확도 향상을 위해 여러 통계적 보정기법을 적용하고 있다. 국내의 경우 기상청에서 가뭄전망을 목적으로 2011년에 수치예보모델을 이용하여 가뭄전망정보를 생산한바 있으나, 전망정보의 불확실성 문제로 가뭄예보에 활용하는데 한계가 있어 이를 개선할 수 있는 기술개발이 요구되는 실정이다. 본 연구에서는 기후예측자료를 이용하여 가뭄전망정보 생산기술을 개발하고 정확도 개선을 위해 베이지안 기법을 연계하였다. GloSea5 (Global Seasonal forecast model 5) 장기예보자료를 이용하였으며, 베이지안 기법을 통해 과거 관측자료에 대한 사전분포, 모델의 전망정보로부터 우도함수를 유도하여 최종 사후분포를 추정하였다. 베이지안 기법 적용 전 후에 따른 가뭄지수를 산정하였다. 관측자료 기반의 가뭄지수와의 비교분석을 통해 선행기간 및 계절별 가뭄예측 성능을 평가하였으며, 실제 가뭄기간 동안에 가뭄의 재현성을 지역별로 분석하였다. 장기예보자료만을 활용한 기존 가뭄전망에서는 관측 자료가 포함된 1개월 전망에서도 불확실성이 매우 높았지만 베이지안 기법 적용으로 가뭄전망의 정확도가 크게 개선되었다. 특히, 1, 2개월 전망의 시계열 가뭄지수가 관측기반의 가뭄지수의 거동과 매우 유사하게 나타났으며, 지역별로도 베이지안 기법 적용시 실제 가뭄피해 상황을 적절히 재현하는 것으로 나타났다. 국내 가뭄예보에 있어 기후예측정보를 단순활용하기 보다는 베이지안과 같은 통계적 보정기법을 이용하여 가뭄전망정보를 생산하는 것이 바람직하며, 본 연구에서는 가뭄예보업무에 활용될 수 있도록 베이지안 기법에 대한 검증 및 평가를 지속적으로 수행할 계획이다.

• Aerospace Engineering and Technology
• /
• v.11 no.1
• /
• pp.169-177
• /
• 2012

This paper considers the explicit guidance algorithm to determine the closed-loop guidance law applicable to upper stages of a given space launch vehicle. It has the advantage of very simple forms derived from the flat earth assumption, which is appropriate for its on-board application. However the simple time-to-go prediction equation produces the degraded guidance performance of the launcher because of its inaccuracy. To overcome the problem, the elaborate prediction equations, which have been employed in Saturn and H-II, are attempted here. Finally, the simulation results show that the simple guidance approach requires the more accurate time-to-go prediction and gravity integrals for its broad application.

• Computational Structural Engineering
• /
• v.10 no.2
• /
• pp.161-169
• /
• 1997

It is known that tire plays an important role to the dynamic performances of a vehicle such as noise, vibration, ride and handling. Therefore, force and moment measurements have been a part of the traditional tire engineering process. In this paper, a computational analysis technique has been explored. A FE model is made to simulate inflation, vertical load due to the vehicle weight, and the slow rolling of a radial tire. A rigid surface with Coulomb friction is included in the model to simulate the slow rolling contact. The tire slip during the in-plane motion of the rigid surface is calculated. Results are presented for both lateral and vertical loads, as well as straight ahead free rolling. The calculated and measured tire slips are in good correlation. A Study on slow Rolling Tire for perdiction of tire Forces and Moments.