• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권5호
• /
• pp.1590-1609
• /
• 2021

Laughter is one of the most important nonverbal sound that human generates. It is a means for expressing his emotions. The acoustic and contextual features of this specific sound are different from those of speech and many difficulties arise during their modeling process. During this work, we propose an audio laughter generation system based on unsupervised generative models: the autoencoder (AE) and its variants. This procedure is the association of three main sub-process, (1) the analysis which consist of extracting the log magnitude spectrogram from the laughter database, (2) the generative models training, (3) the synthesis stage which incorporate the involvement of an intermediate mechanism: the vocoder. To improve the synthesis quality, we suggest two hybrid models (LSTM-VAE, GRU-VAE and CNN-VAE) that combine the representation learning capacity of variational autoencoder (VAE) with the temporal modelling ability of a long short-term memory RNN (LSTM) and the CNN ability to learn invariant features. To figure out the performance of our proposed audio laughter generation process, objective evaluation (RMSE) and a perceptual audio quality test (listening test) were conducted. According to these evaluation metrics, we can show that the GRU-VAE outperforms the other VAE models.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제28권1호
• /
• pp.1-21
• /
• 2024

This review comprehensively explores the evolution, theoretical underpinnings, variations, and applications of diffusion models. Originating as a generative framework, diffusion models have rapidly ascended to the forefront of machine learning research, owing to their exceptional capability, stability, and versatility. We dissect the core principles driving diffusion processes, elucidating their mathematical foundations and the mechanisms by which they iteratively refine noise into structured data. We highlight pivotal advancements and the integration of auxiliary techniques that have significantly enhanced their efficiency and stability. Variants such as bridges that broaden the applicability of diffusion models to wider domains are introduced. We put special emphasis on the ability of diffusion models as a crucial foundation model, with modalities ranging from image, 3D assets, and video. The role of diffusion models as a general foundation model leads to its versatility in many of the downstream tasks such as solving inverse problems and image editing. Through this review, we aim to provide a thorough and accessible compendium for both newcomers and seasoned researchers in the field.

• 한국지능시스템학회논문지
• /
• 제24권3호
• /
• pp.233-238
• /
• 2014

독창성은 예술작품 뿐만아니라 장신구, 생활용품 등 다양한 디자인에서 요구된다. 창조적 디자인 작업에 참신한 모티프를 얻기 위해 변환생성 디자인 기법이 활용될 수 있다. 이 논문에서는 유전자 알고리즘을 이용하여 독특한 반지 모델을 만들어 내는 변환생성 디자인 방법을 제안한다. 후보해 표현방법, 연산자, 평가함수 관점에서 제안된 알고리즘에 대해서 소개한다. 제안한 방법은 고객이 자신의 반지 모양에 대한 취향을 선택하도록 하고, 취향을 평가에 반영하도록 하면서 여러 가지 반지 모델을 만들어서 추천하도록 한다. 반지 모델은 3차원 입체로 표현되기 때문에, 고객이 최종적으로 선택한 모델을 3차원 프린터를 통해서 실물로 제작될 수 있다.

• 한국과학교육학회지
• /
• 제16권1호
• /
• pp.13-34
• /
• 1996

The purpose of this study is to develop appropriate science teaching models which can be applied effectively to relevant situations. Five science teaching models; cognitive conflict teaching models, generative teaching model, learning cycle teaching model, hypothesis verification teaching model and discovery teaching model, were identified from the existing models. The teaching models were modified and in primary and secondary students using a nonequivalent pretest-posttest control group design. Major findings of this study were as follows: 1. For teaching science concepts, three teaching models were found more effective; cognitive conflict teaching model, generative teaching model and discovery teaching model. 2. For teaching inquiry skills, two teaching models were found more effective; learning cycle teaching model and hypothesis verification teaching model. 3. For teaching scientific attitudes, two teaching models were found more effective; learning cycle teaching models and discovery teaching model. Each teaching model requires specific learning environment. It is strongly suggested that teachers should select a suitable teaching model carefully after evaluating the learning environment including teacher and student variables, learning objectives and curricular materials.

• International Journal of Internet, Broadcasting and Communication
• /
• 제11권4호
• /
• pp.37-42
• /
• 2019

In this paper, we explore the details of three classic data augmentation methods and two generative model based oversampling methods. The three classic data augmentation methods are random sampling (RANDOM), Synthetic Minority Over-sampling Technique (SMOTE), and Adaptive Synthetic Sampling (ADASYN). The two generative model based oversampling methods are Conditional Generative Adversarial Network (CGAN) and Wasserstein Generative Adversarial Network (WGAN). In imbalanced data, the whole instances are divided into majority class and minority class, where majority class occupies most of the instances in the training set and minority class only includes a few instances. Generative models have their own advantages when they are used to generate more plausible samples referring to the distribution of the minority class. We also adopt CGAN to compare the data augmentation performance with other methods. The experimental results show that WGAN-based oversampling technique is more stable than other approaches (RANDOM, SMOTE, ADASYN and CGAN) even with the very limited training datasets. However, when the imbalanced ratio is too small, generative model based approaches cannot achieve satisfying performance than the conventional data augmentation techniques. These results suggest us one of future research directions.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권3호
• /
• pp.186-199
• /
• 2013

We pose pattern classification as a density estimation problem where we consider mixtures of generative models under partially labeled data setups. Unlike traditional approaches that estimate density everywhere in data space, we focus on the density along the decision boundary that can yield more discriminative models with superior classification performance. We extend our earlier work on the recursive estimation method for discriminative mixture models to semi-supervised learning setups where some of the data points lack class labels. Our model exploits the mixture structure in the functional gradient framework: it searches for the base mixture component model in a greedy fashion, maximizing the conditional class likelihoods for the labeled data and at the same time minimizing the uncertainty of class label prediction for unlabeled data points. The objective can be effectively imposed as individual mixture component learning on weighted data, hence our mixture learning typically becomes highly efficient for popular base generative models like Gaussians or hidden Markov models. Moreover, apart from the expectation-maximization algorithm, the proposed recursive estimation has several advantages including the lack of need for a pre-determined mixture order and robustness to the choice of initial parameters. We demonstrate the benefits of the proposed approach on a comprehensive set of evaluations consisting of diverse time-series classification problems in semi-supervised scenarios.

• 한국융합학회논문지
• /
• 제13권1호
• /
• pp.13-21
• /
• 2022

최근 딥러닝 기반의 자연어처리 연구는 사전 훈련된 언어 모델을 통해 대부분의 자연어처리 분야에서 우수한 성능을 보인다. 특히 오토인코더 (auto-encoder) 기반의 언어 모델은 다양한 한국어 이해 분야에서 뛰어난 성능과 쓰임을 증명하고 있다. 그러나 여전히 디코더 (decoder) 기반의 한국어 생성 모델은 간단한 문장 생성 과제에도 어려움을 겪고 있으며, 생성 모델이 가장 일반적으로 쓰이는 대화 분야에서의 세부 연구와 학습 가능한 데이터가 부족한 상황이다. 따라서 본 논문은 한국어 생성 모델을 위한 멀티턴 대화 데이터를 구축하고 전이 학습을 통해 생성 모델의 대화 능력을 개선하여 성능을 비교 분석한다. 또한, 검색 모델을 통해 외부 지식 정보에서 추천 응답 후보군을 추출하여 모델의 부족한 대화 생성 능력을 보완하는 방법을 제안한다.

• 문화기술의 융합
• /
• 제8권6호
• /
• pp.867-871
• /
• 2022

최근 빠른 유행의 변화 속에서 디자인의 변화는 패션기업의 매출에 큰 영향을 미치기 때문에 기업들은 신제품디자인 선택에 신중할 수밖에 없다. 최근 인공지능 분야의 발달에 따라 패션시장에서도 소비자들의 선호도를 높이기 위해 다양한 기계학습을 많이 활용하고 있다. 우리는 선호도와 같은 추상적인 개념을 수치화함으로써 신제품 개발에 신뢰성을 높이는 부분에 기여하고자 한다. 이를 위해 3가지 적대적 생성 신경망(Generative adversial netwrok, GAN)을 통하여 기존에 없는 새로운 이미지를 생성하고, 미리 훈련된 합성곱 신경망(Convolution neural networkm, CNN)을 이용하여 선호도라는 추상적인 개념을 수치화시켜 비교하였다. 심층 컨볼루션 적대적 생성 신경망(Deep convolutional generative adversial netwrok, DCGAN), 점진적 성장 적대적 생성 신경망(Progressive growing generative adversial netwrok, PGGAN), 이중 판별기 적대적 생성 신경망(Dual Discriminator generative adversial netwrok, D2GAN)의 3가지 방법을 통해 새로운 이미지를 생성하였고, 판매량이 높았던 제품으로 훈련된 합성곱 신경망으로 유사도를 비교, 측정하였다. 측정된 유사도의 정도를 선호도로 간주하였으며 실험 결과 D2GAN이 DCGAN, PGGAN에 비해 상대적으로 높은 유사도를 보여주었다.

• International journal of advanced smart convergence
• /
• 제13권2호
• /
• pp.130-135
• /
• 2024

The Stable diffsuion AI tool is popular among designers because of its flexible and powerful image generation capabilities. However, due to the diversity of its AI models, it needs to spend a lot of time testing different AI models in the face of different design plans, so choosing a suitable general AI model has become a big problem at present. In this paper, by comparing the AI images generated by two different Stable diffsuion models, the advantages and disadvantages of each model are analyzed from the aspects of the matching degree of the AI image and the prompt, the color composition and light composition of the image, and the general AI model that the generated AI image has an aesthetic sense is analyzed, and the designer does not need to take cumbersome steps. A satisfactory AI image can be obtained. The results show that Playground V2.5 model can be used as a general AI model, which has both aesthetic and design sense in various style design requirements. As a result, content designers can focus more on creative content development, and expect more groundbreaking technologies to merge generative AI with content design.

• 방송공학회논문지
• /
• 제28권3호
• /
• pp.275-284
• /
• 2023

유전자 발현 데이터는 질병의 예후 예측, 약물 반응성 예측 등 질병에 대한 이해와 정밀 의료 실현을 위한 연구들에 활용될 수 있지만 충분한 양의 데이터를 수집하는 데 많은 비용적 문제가 있다. 본 논문에서는 Conditional VAE에 기반한 유전자 발현 데이터 생성 모델을 제안하였다. 이전 연구인 WGAN-GP기반의 유전자 발현 생성 모델과 정형 데이터 생성 모델인 CTGAN, TVAE와 비교하여 본 논문의 Conditional VAE기반 모델이 생물학적, 통계학적으로 더 유의미한 합성 데이터를 생성할 수 있음을 보였다.