• 지능정보연구
• /
• 제29권4호
• /
• pp.165-184
• /
• 2023

최근 구글의 BERT, OpenAI의 GPT 등, 언어모델(Language Model)을 사용한 비정형 텍스트 데이터에 대한 딥러닝(Deep Learning) 분석이 다양한 응용에서 괄목할 성과를 나타내고 있다. 대부분의 언어모델은 사전학습 데이터로부터 범용적인 언어정보를 학습하고, 이후 미세 조정(Fine-Tuning) 과정을 통해 다운스트림 태스크(Downstream Task)에 맞추어 갱신되는 방식으로 사용되고 있다. 하지만 최근 이러한 언어모델을 사용하는 과정에서 프라이버시가 침해될 수 있다는 우려가 제기되고 있다. 즉 데이터 소유자가 언어모델의 미세 조정을 수행하기 위해 다량의 데이터를 모델 소유자에게 제공하는 과정에서 데이터의 프라이버시가 침해될 수 있으며, 반대로 모델 소유자가 모델 전체를 데이터 소유자에게 공개하면 모델의 구조 및 가중치가 공개되어 모델의 프라이버시가 침해될 수 있다는 것이다. 이러한 상황에서 프라이버시를 보호하며 언어모델의 미세 조정을 수행하기 위해 최근 오프사이트 튜닝(Offsite Tuning)의 개념이 제안되었으나, 해당 연구는 제안 방법론을 텍스트 분류 모델에 적용하는 구체적인 방안을 제시하지 못했다는 한계를 갖는다. 이에 본 연구에서는 한글 문서에 대한 다중 분류 미세 조정 수행 시, 모델과 데이터의 프라이버시를 보호하기 위해 분류기를 추가한 오프사이트 튜닝을 적용하는 구체적인 방법을 제시한다. 제안 방법론의 성능을 평가하기 위해 AIHub에서 제공하는 ICT, 전기, 전자, 기계, 그리고 의학 총 5개의 대분야로 구성된 약 20만건의 한글 데이터에 대해 실험을 수행한 결과, 제안하는 플러그인 모델이 제로 샷 모델 및 오프사이트 모델에 비해 분류 정확도 측면에서 우수한 성능을 나타냄을 확인하였다.

• 전기전자학회논문지
• /
• 제27권3호
• /
• pp.313-318
• /
• 2023

본 논문은 Vision Transformer를 기반으로 하는 Video Classification의 성능을 개선하는 방법으로 fine-tuning를 적용한 신경망을 제안한다. 최근 딥러닝 기반 실시간 비디오 영상 분석의 필요성이 대두되고 있다. Image Classification에 사용되는 기존 CNN 모델의 특징상 연속된 Frame에 대한 연관성을 분석하기 어렵다는 단점이 있다. 이와 같은 문제를 Attention 메커니즘이 적용된 Vistion Transformer와 Non-local 신경망 모델을 비교 분석하여 최적의 모델을 찾아 해결하고자 한다. 또한, 전이 학습 방법으로 fine-tuning의 다양한 방법을 적용하여 최적의 fine-tuning 신경망 모델을 제안한다. 실험은 UCF101 데이터셋으로 모델을 학습시킨 후, UTA-RLDD 데이터셋에 전이 학습 방법을 적용하여 모델의 성능을 검증하였다.

• 한국멀티미디어학회논문지
• /
• 제24권7호
• /
• pp.849-859
• /
• 2021

Prostate cancer (PCa) is a fatal disease that occurs in men. In general, PCa cells are found in the prostate gland. Early diagnosis is the key to prevent the spreading of cancers to other parts of the body. In this case, deep learning-based systems can detect and distinguish histological patterns in microscopy images. The histological grades used for the analysis were benign, grade 3, grade 4, and grade 5. In this study, we attempt to use transfer learning and fine-tuning methods as well as different model architectures to develop and compare the models. We implemented MobileNet, ResNet50, and DenseNet121 models and used three different strategies of freezing layers techniques of fine-tuning, to get various pre-trained weights to improve accuracy. Finally, transfer learning using MobileNet with the half-layer frozen showed the best results among the nine models, and 90% accuracy was obtained on the test data set.

• 한국IT서비스학회지
• /
• 제19권5호
• /
• pp.83-91
• /
• 2020

Language Models such as BERT has been an important factor of deep learning-based natural language processing. Pre-training the transformer-based language models would be computationally expensive since they are consist of deep and broad architecture and layers using an attention mechanism and also require huge amount of data to train. Hence, it became mandatory to do fine-tuning large pre-trained language models which are trained by Google or some companies can afford the resources and cost. There are various techniques for fine tuning the language models and this paper examines three techniques, which are data augmentation, tuning the hyper paramters and partly re-constructing the neural networks. For data augmentation, we use no-answer augmentation and back-translation method. Also, some useful combinations of hyper parameters are observed by conducting a number of experiments. Finally, we have GRU, LSTM networks to boost our model performance with adding those networks to BERT pre-trained model. We do fine-tuning the pre-trained korean-based language model through the methods mentioned above and push the F1 score from baseline up to 89.66. Moreover, some failure attempts give us important lessons and tell us the further direction in a good way.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권12호
• /
• pp.4816-4834
• /
• 2020

This paper proposes transfer learning and fine-tuning techniques for a deep learning model to detect three distinct brain tumors from Magnetic Resonance Imaging (MRI) scans. In this work, the recent YOLOv4 model trained using a collection of 3064 T1-weighted Contrast-Enhanced (CE)-MRI scans that were pre-processed and labeled for the task. This work trained with the partial 29-layer YOLOv4-Tiny and fine-tuned to work optimally and run efficiently in most platforms with reliable performance. With the help of transfer learning, the model had initial leverage to train faster with pre-trained weights from the COCO dataset, generating a robust set of features required for brain tumor detection. The results yielded the highest mean average precision of 93.14%, a 90.34% precision, 88.58% recall, and 89.45% F1-Score outperforming other previous versions of the YOLO detection models and other studies that used bounding box detections for the same task like Faster R-CNN. As concluded, the YOLOv4-Tiny can work efficiently to detect brain tumors automatically at a rapid phase with the help of proper fine-tuning and transfer learning. This work contributes mainly to assist medical experts in the diagnostic process of brain tumors.

• 말소리와 음성과학
• /
• 제15권3호
• /
• pp.83-88
• /
• 2023

This study investigates the fine-tuning of large-scale Automatic Speech Recognition (ASR) models, specifically OpenAI's Whisper model, for domain-specific applications using the KsponSpeech dataset. The primary research questions address the effectiveness of targeted lexical item emphasis during fine-tuning, its impact on domain-specific performance, and whether the fine-tuned model can maintain generalization capabilities across different languages and environments. Experiments were conducted using two fine-tuning datasets: Set A, a small subset emphasizing specific lexical items, and Set B, consisting of the entire KsponSpeech dataset. Results showed that fine-tuning with targeted lexical items increased recognition accuracy and improved domain-specific performance, with generalization capabilities maintained when fine-tuned with a smaller dataset. For noisier environments, a trade-off between specificity and generalization capabilities was observed. This study highlights the potential of fine-tuning using minimal domain-specific data to achieve satisfactory results, emphasizing the importance of balancing specialization and generalization for ASR models. Future research could explore different fine-tuning strategies and novel technologies such as prompting to further enhance large-scale ASR models' domain-specific performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권3호
• /
• pp.570-590
• /
• 2024

Breast cancer ranks among the most prevalent forms of malignancy and foremost cause of death by cancer worldwide. It is not preventable. Early and precise detection is the only remedy for lowering the rate of mortality and improving the probability of survival for victims. In contrast to present procedures, thermography aids in the early diagnosis of cancer and thereby saves lives. But the accuracy experiences detrimental impact by low sensitivity for small and deep tumours and the subjectivity by physicians in interpreting the images. Employing deep learning approaches for cancer detection can enhance the efficacy. This study explored the utilization of thermography in early identification of breast cancer with the use of a publicly released dataset known as the DMR-IR dataset. For this purpose, we employed a novel approach that entails the utilization of a pre-trained MobileNetV2 model and fine tuning it through transfer learning techniques. We created three models using MobileNetV2: one was a baseline transfer learning model with weights trained from ImageNet dataset, the second was a fine-tuned model with an adaptive learning rate, and the third utilized early stopping with callbacks during fine-tuning. The results showed that the proposed methods achieved average accuracy rates of 85.15%, 95.19%, and 98.69%, respectively, with various performance indicators such as precision, sensitivity and specificity also being investigated.

• 한국정보기술학회 영문논문지
• /
• 제10권1호
• /
• pp.45-56
• /
• 2020

Despite extensive research, performance enhancement of keyphrase (KP) extraction remains a challenging problem in modern informatics. Recently, deep learning-based supervised approaches have exhibited state-of-the-art accuracies with respect to this problem, and several of the previously proposed methods utilize Bidirectional Encoder Representations from Transformers (BERT)-based language models. However, few studies have investigated the effective application of BERT-based fine-tuning techniques to the problem of KP extraction. In this paper, we consider the aforementioned problem in the context of scientific articles by investigating the fine-tuning characteristics of two distinct BERT models - BERT (i.e., base BERT model by Google) and SciBERT (i.e., a BERT model trained on scientific text). Three different datasets (WWW, KDD, and Inspec) comprising data obtained from the computer science domain are used to compare the results obtained by fine-tuning BERT and SciBERT in terms of KP extraction.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권1호
• /
• pp.147-165
• /
• 2021

In this work, we empirically evaluated the efficiency of the recent EfficientNetB0 model to identify and diagnose malaria parasite infections in blood smears. The dataset used was collected and classified by relevant experts from the Lister Hill National Centre for Biomedical Communications (LHNCBC). We prepared our samples with minimal image transformations as opposed to others, as we focused more on the feature extraction capability of the EfficientNetB0 baseline model. We applied transfer learning to increase the initial feature sets and reduced the training time to train our model. We then fine-tuned it to work with our proposed layers and re-trained the entire model to learn from our prepared dataset. The highest overall accuracy attained from our evaluated results was 94.70% from fifty epochs and followed by 94.68% within just ten. Additional visualization and analysis using the Gradient-weighted Class Activation Mapping (Grad-CAM) algorithm visualized how effectively our fine-tuned EfficientNetB0 detected infections better than other recent state-of-the-art DCNN models. This study, therefore, concludes that when fine-tuned, the recent EfficientNetB0 will generate highly accurate deep learning solutions for the identification of malaria parasites in blood smears without the need for stringent pre-processing, optimization, or data augmentation of images.

• Journal of Semiconductor Engineering
• /
• 제2권1호
• /
• pp.125-129
• /
• 2021

As the interest of deep learning, techniques to control the color of images in image processing field are evolving together. However, there is no clear standard for color, and it is not easy to find a way to represent only the color itself like the color-palette. In this paper, we propose a novel color palette extraction system by chroma fine-tuning with reinforcement learning. It helps to recognize the color combination to represent an input image. First, we use RGBY images to create feature maps by transferring the backbone network with well-trained model-weight which is verified at super resolution convolutional neural networks. Second, feature maps are trained to 3 fully connected layers for the color-palette generation with a generative adversarial network (GAN). Third, we use the reinforcement learning method which only changes chroma information of the GAN-output by slightly moving each Y component of YCbCr color gamut of pixel values up and down. The proposed method outperforms existing color palette extraction methods as given the accuracy of 0.9140.