• 음성과학
• /
• 제10권1호
• /
• pp.71-84
• /
• 2003

In the acoustic modeling for large vocabulary speech recognition, a sparse data problem caused by a huge number of context-dependent (CD) models usually leads the estimated models to being unreliable. In this paper, we develop a new clustering method based on the C45 decision-tree learning algorithm that effectively encapsulates the CD modeling. The proposed scheme essentially constructs a supervised decision rule and applies over the pre-clustered triphones using the C45 algorithm, which is known to effectively search through the attributes of the training instances and extract the attribute that best separates the given examples. In particular, the data driven method is used as a clustering algorithm while its result is used as the learning target of the C45 algorithm. This scheme has been shown to be effective particularly over the database of low unknown-context ratio in terms of recognition performance. For speaker-independent, task-independent continuous speech recognition task, the proposed method reduced the percent accuracy WER by 3.93% compared to the existing rule-based methods.

• Journal of Information Processing Systems
• /
• 제19권2호
• /
• pp.232-239
• /
• 2023

Aiming at the poor compression quality of traditional image compression coding (ICC) algorithm, a multi-description ICC algorithm based on depth learning is put forward in this study. In this study, first an image compression algorithm was designed based on multi-description coding theory. Image compression samples were collected, and the measurement matrix was calculated. Then, it processed the multi-description ICC sample set by using the convolutional self-coding neural system in depth learning. Compressing the wavelet coefficients after coding and synthesizing the multi-description image band sparse matrix obtained the multi-description ICC sequence. Averaging the multi-description image coding data in accordance with the effective single point's position could finally realize the compression coding of multi-description images. According to experimental results, the designed algorithm consumes less time for image compression, and exhibits better image compression quality and better image reconstruction effect.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• 지능정보연구
• /
• 제29권3호
• /
• pp.167-183
• /
• 2023

최근 추천시스템 분야에서는 희소한 데이터를 효과적으로 모델링하기 위한 다양한 연구가 진행되고 있다. GLocal-K(Global and Local Kernels for Recommender Systems)는 그중 하나의 연구로 전역 커널과 지역 커널을 결합하여 데이터의 전역적인 패턴과 개별 사용자의 특성을 모두 고려해 사용자 맞춤형 추천을 제공하는 모델이다. 하지만 GLocal-K는 커널 트릭을 사용하기 때문에 매우 희소한 데이터에서 성능이 떨어지고 부가 정보를 사용하지 않아 새로운 사용자나 아이템에 대한 추천을 제공하는 데 어려움이 있다. 본 논문에서는 이러한 GLocal-K의 단점을 극복하기 위해 EASE(Embarrassingly Shallow Autoencoders for Sparse Data) 모델과 부가 정보를 활용한 GEase-K(Global and EASE kernels for Recommender Systems) 모델을 제안한다. 우선 GLocal-K의 지역 커널 대신 EASE를 활용하여 매우 희소한 데이터에서 추천 성능을 높이고자 하였다. EASE는 단순한 선형 연산 구조로 이루어져 있지만, 규제화와 아이템 간 유사도 학습을 통해 매우 희소한 데이터에서 높은 성능을 내는 오토인코더이다. 다음으로 Cold Start 완화를 위해 부가 정보를 활용하였다. 학습 과정에서 부가 정보를 추가하기 위해 조건부 오토인코더 구조를 적용하였으며 이를 통해 사용자-아이템 간의 유사성을 더 잘 파악할 수 있도록 하였다. 결론적으로 GEase-K는 선형 구조와 비선형 구조의 결합, 부가 정보의 활용을 통해 매우 희소한 데이터와 Cold Start 상황에서 강건한 모습을 보인다. 실험 결과, GEase-K는 매우 희소한 GoodReads, ModCloth 데이터 세트에서 RMSE, MAE 평가 지표 기준 GLocal-K 보다 높은 성능을 보였다. 또한 GoodReads, ModCloth 데이터 세트를 4개의 집단으로 나누어 실험한 Cold Start 실험에서도 GLocal-K 대비 Cold Start 상황에서 좋은 성능을 보였다.

• 대한영상의학회지
• /
• 제85권2호
• /
• pp.345-362
• /
• 2024

자기공명영상(이하 MRI)은 복부 영상에서 국소 병변의 감지와 특성을 찾을 수 있는 것 때문에 중요한 역할을 한다. 그러나 MRI 검사에 상대적으로 긴 검사 시간과 호흡 유지 기법에서 움직임 관리와 같은 몇 가지 힘든 요인이 있다. 최근에는 검사 시간을 줄이면서 적절한 이미지 품질을 유지하는 기법인 평행 이미징, 압축 감지(compressed sensing) 및 최첨단 딥 러닝(deep learning) 기술이 등장하여 문제 해결 전략을 가능하게 하고 있다. 또한, 역동적 조영증강 영상에서 자유 호흡 기법은, 추가 차원(extra-dimensional)-부피 보간 호흡 유지 검사(volumetric interpolated breath-hold examination) 및 황금 각도 방사형 희소 병렬(golden-angle radial sparse parallel), 간 가속 볼륨 획득(liver acceleration volume acquisition) 스타와 같은, 심한 호흡곤란이나 마취 중인 환자에게서 복부 MRI를 시행하는 것을 돕는다. 이 임상화보에서는 시간을 줄이면서도 이미지 품질을 유지하기 위한 다양한 고급 복부 MRI 기술과 역동적 영상을 위한 자유 호흡 기술을 제시하고 또한 이를 통한 예시들을 보여주고자 한다. 이러한 첨단 기법들의 고찰은 적용된 시퀀스의 적절한 해석에 도움을 줄 것이다.

• Journal of Information Science Theory and Practice
• /
• 제1권1호
• /
• pp.10-26
• /
• 2013

Domain transfer is a widely recognized problem for machine learning algorithms because models built upon one data domain generally do not perform well in another data domain. This is especially a challenge for tasks such as opinion classification, which often has to deal with insufficient quantities of labeled data. This study investigates the feasibility of self-training in dealing with the domain transfer problem in opinion classification via leveraging labeled data in non-target data domain(s) and unlabeled data in the target-domain. Specifically, self-training is evaluated for effectiveness in sparse data situations and feasibility for domain adaptation in opinion classification. Three types of Web content are tested: edited news articles, semi-structured movie reviews, and the informal and unstructured content of the blogosphere. Findings of this study suggest that, when there are limited labeled data, self-training is a promising approach for opinion classification, although the contributions vary across data domains. Significant improvement was demonstrated for the most challenging data domain-the blogosphere-when a domain transfer-based self-training strategy was implemented.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1403-1417
• /
• 2019

With the rapid growth of multimedia data, research on cross-media feature learning has significance in many applications, such as multimedia search and recommendation. Existing methods are sensitive to noise and edge information in multimedia data. In this paper, we propose a semi-supervised method for cross-media feature learning by means of $L_{2,q}$ norm to improve the performance of cross-media retrieval, which is more robust and efficient than the previous ones. In our method, noise and edge information have less effect on the results of cross-media retrieval and the dynamic patch information of multimedia data is employed to increase the accuracy of cross-media retrieval. Our method can reduce the interference of noise and edge information and achieve fast convergence. Extensive experiments on the XMedia dataset illustrate that our method has better performance than the state-of-the-art methods.

• 한국산업융합학회 논문집
• /
• 제24권1호
• /
• pp.1-7
• /
• 2021

In this paper, we presented the experimental basis for the theoretical background and robustness of the Convolutional Neural Network for object recognition based on artificial intelligence. An experimental result was performed to visualize the weighting filters and feature maps for each layer to determine what characteristics CNN is automatically generating. experimental results were presented on the trend of learning error and identification error rate by checking the relevance of the weight filter and feature map for learning error and identification error. The weighting filter and characteristic map are presented as experimental results. The automatically generated characteristic quantities presented the results of error rates for moving and rotating robustness to geometric changes.

• Journal of Information Processing Systems
• /
• 제17권6호
• /
• pp.1057-1070
• /
• 2021

Case-related news filtering is crucial in legal text mining and divides news into case-related and case-unrelated categories. Because case-related news originates from various fields and has different writing styles, it is difficult to establish complete filtering rules or keywords for data collection. In addition, the labeled corpus for case-related news is sparse; therefore, to train a high-performance classification model, it is necessary to annotate the corpus. To address this challenge, we propose topic-enhanced positive-unlabeled learning, which selects positive and negative samples guided by topics. Specifically, a topic model based on a variational autoencoder (VAE) is trained to extract topics from unlabeled samples. By using these topics in the iterative process of positive-unlabeled (PU) learning, the accuracy of identifying case-related news can be improved. From the experimental results, it can be observed that the F1 value of our method on the test set is 1.8% higher than that of the PU learning baseline model. In addition, our method is more robust with low initial samples and high iterations, and compared with advanced PU learning baselines such as nnPU and I-PU, we obtain a 1.1% higher F1 value, which indicates that our method can effectively identify case-related news.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권3호
• /
• pp.792-812
• /
• 2022

Multi-agent systems often need to achieve the goal of learning more effectively for a task through coordination. Although the introduction of deep learning has addressed the state space problems, multi-agent learning remains infeasible because of the joint action spaces. Large-scale joint action spaces can be sparse according to implicit or explicit coordination structure, which can ensure reasonable coordination action through the coordination structure. In general, the multi-agent system is dynamic, which makes the relations among agents and the coordination structure are dynamic. Therefore, the explicit coordination structure can better represent the coordinative relationship among agents and achieve better coordination between agents. Inspired by the maximization of social group utility, we dynamically construct a factor graph as an explicit coordination structure to express the coordinative relationship according to the utility among agents and estimate the joint action values based on the local utility transfer among factor graphs. We present the application of such techniques in the scenario of multiple intelligent vehicle systems, where state space and action space are a problem and have too many interactions among agents. The results on the multiple intelligent vehicle systems demonstrate the efficiency and effectiveness of our proposed methods.