Lithography is the process of transferring the geometric patterns from the masks to the resist material on the semiconductor. It is a very important part of VLSI fabrication that is critical when it comes to the efficient functioning of circuits. Many state-of-the-art methods use Machine Learning (ML) to identify lithography patterns that can cause issues in the future as these algorithms can predict defects in patterns which the machine has not encountered before. This paper focuses on the need for Machine Learning in the lithography process, and the various algorithms used like Support Vector Machines (SVM), Artificial Neural Networks (ANN) and Convolutional Neural Networks (CNN). There are multiple applications including Hotspot detection, Optical Proximity Correction (OPC), Sub Resolution Assist Feature (SRAF), Phase Shift Masks (PSM), and Resist Modelling. The major issue faced by Machine Learning algorithms is that of false positives. It can be reduced by utilizing the Gaussian process after initial detection. This paper explores the current methods of how machine learning is used to improve VLSI fabrication and compares strengths and weaknesses of the methods.