Semi Supervised Learning Approaches for Semiconductors Defect Classification

In recent years, deep neural networks have emerged as an evolved branch of Machine Learning (ML). The methods have drawn the research community’s attention as a powerful tool in data-driven tasks. Therefore, recent researchers are shedding light on empowering the networks’ topologies and architectures. Sufficient data is also a key element in the training empirical success. The main issue relies on the labeled data scarcity. In fact, data annotation can be a time-consuming and costly process. To address this issue, Semi Supervised Learning (SSL) is a Machine Learning (ML) technique that proposes effective approaches to exploit the unlabeled data availability. The purpose of this thesis is to implement SSL methods applied to Semiconductors Defect Classification scenario.