On Detecting Rare Patterns 08/2018 - 05/2019 Advisor: Dr. Christino Tamon • Implemented from scratch the model and replicated the experiments in Finite Sample Complexity of Rare Pattern Anomaly Detection • Analyzed and identified the model weaknesses, creating an improved model to accommodate imbalanced datasets and outperform other popular outlier detection algorithms, such as GMM, PPCA, LSA, LOF, one−class SVM, and Isolation Forest on categorical datasets, such as shuttle, wine-quality, ann-thyroid, mammography, yeast, and covertype • Calculated AUC scores to evaluate the model performance