A Comparative Study of Transfer Learning and Fine-Tuning Method on Deep Learning Models for Wayang Dataset Classification

Abstract

Deep Learning is an essential technique in the classification problem in machine learning based on artificial neural networks. The general issue in deep learning is data-hungry, which require a plethora of data to train some model. Wayang is a shadow puppet art theater from Indonesia, especially in the Javanese culture. It has several indistinguishable characters. In this paper, We tried proposing some steps and techniques on how to classify the characters and handle the issue on a small wayang dataset by using model selection, transfer learning, and fine-tuning to obtain efficient and precise accuracy on our classification problem. The research used 50 images for each class and a total of 24 wayang characters classes. We collected and implemented various architectures from the initial version of deep learning to the latest proposed model and their state-of-art. The transfer learning and fine-tuning method showed a significant increase in accuracy, validation accuracy. By using Transfer Learning, it was possible to design the deep learning model with good classifiers within a short number of times on a small dataset. It performed 100% on their training on both EfficientNetB0 and MobileNetV3-small. On validation accuracy, gave 98.33% and 98.75%, respectively.

Classification of Spotted Disease on Sugarcane Leaf Image Using Convolutional Neural Network Algorithm

“…Dalam pembelajaran mendalam, pembelajaran transfer mengacu pada penggunaan kembali jaringan yang telah dilatih sebelumnya pada pekerjaan baru. Parameter bobot yang dihasilkan atau model pra-pelatihan dari domain sumber selanjutnya digunakan dalam domain target [15].…”

The effects of artificial intelligence on the Kenyan society

“…Currently, AI is being used in areas such as diagnosis in medicine, translation, autonomous vehicles, and risk assessment and management of investments in the financial sector. Machine learning usually requires little to no interaction with humans [10]. AI solves these areas' problems usually by breaking the problem solving process into subphases [11].…”

“…AI solves these areas' problems usually by breaking the problem solving process into subphases [11]. Additionally, a tiny dataset can be used to rapidly create a deep learning model with high-quality classifiers [10]. An eleven-person "Blockchain and AI Taskforce" has been assembled by Kenya's Ministry of Information, Communication, and Technology (ICT) to investigate how these technologies might be used for the country's development.…”