Display omitted •A machine learning based strategy was proposed to improve the processing fidelity of the vat polymerization process by automatically tuning the distribution of localized grey-pixels of different intensities.•A chessboard pattern based data generation and training strategy was proposed to help the model better understand the impact of curing resulting from different combinations of adjoining greyscale pixel levels.•An automated data processing strategy to help extract and repair data after data collection.•The introducing of CGAN system for increasing the sets of training data. Like many pixel-based additive manufacturing (AM) techniques, digital light processing (DLP) based vat photopolymerization faces the challenge that the square pixel based processing strategy can lead to zigzag edges especially when feature sizes come close to single-pixel levels. Introducing greyscale pixels has been a strategy to smoothen such edges, but it is a challenging task to understand which of the many permutations of projected pixels would give the optimal 3D printing performance. To address this challenge, a novel data acquisition strategy based on machine learning (ML) principles is proposed, and a training routine is implemented to reproduce the smallest shape of an intended 3D printed object. Through this approach, a chessboard patterning strategy is developed along with an automated data refining and augmentation workflow, demonstrating its efficiency and effectiveness by reducing the deviation by around 30%.