Machine Learning Sessional

CSE 472 : Machine Learning Sessional

• Matrix Transformation, Eigen Decomposition & Image Reconstruction using SVD
• 4 basic tasks
• Logistic Regression and AdaBoost for classification
• Worked on 3 datasets for classification task
• Learned about data pre-processing
• Performance measure : Accuracy, Sensitivity, Specificity, Precision, False discovery rate, F1 score
• Measured performance changing boosting rounds
• Feed Forward Network from Scratch
• Activation Function : Relu, Softmax, tanh
• Loss Function : L1 Loss, L2 Loss, Binary Cross Entropy Loss
• 
• Assignment on PCA and EM Algorithm
  • 2D, 3D and 6D data points
  • Estimation of Clusture