Daily AI Problems

You are training your Deep Learning model. After training, you realize the training accuracy is good. However, validation accuracy is poor. What could have been the reason behind it?

When does normalization of features help?

Which of the following is/are true for transformations on data while using tree-based models?

What does "neg_mean_absolute_error" mean?

You are performing Exploratory Data Analysis on your Tabular Dataset. You observe that a column of your dataset has around 10% missing values. Which of the following would be the best alternative to proceed with? (Assume number of rows ~ 10K)

You just trained a ML model on a tabular dataset with an aim of detecting SPAM emails. A False positive would mean ham classified as SPAM whereas False negative would mean SPAM classified as ham. Your default model threshold was 0.5. Will you increase the threshold, decrease it, or keep the same to improve model prediction performance? (Assuming the model is on a learning cycle)

You are given a task of image segmentation. You decide to implement UNet to solve the problem. You observe that the segmentation masks generated by the model after training aren't very accurate ( IoU score or Dice). Which one of the following is a correct inference?