ML Crossword #2

1234567891011121314151617181920
Across
  1. 5. Data used to approximate labels not directly available in a dataset
  2. 8. The number of neurons in a particular layer of a neural network
  3. 9. The model correctly predicted the positive class
  4. 11. A form of regularization useful in training neural networks
  5. 12. An algorithm that implements quantile bucketing on a particular feature in a dataset
  6. 13. A sophisticated gradient descent algorithm that prevents learning from getting stuck in local minima
  7. 14. A full training pass over the entire dataset such that each example has been seen once
  8. 15. Creating a model that matches the training data so closely that the model fails to make correct predictions on new data
  9. 18. The correct answer. Reality
  10. 20. A boundary that separates a space into two subspaces
Down
  1. 1. An open-source math library that provides efficient array operations in Python
  2. 2. A process used, as part of training, to evaluate the quality of a machine learning
  3. 3. The portion of a LSTM cell that regulates the flow of information through the cell
  4. 4. Applying a weight to the downsampled class equal to the factor by which you downsampled
  5. 6. A graph of weight(s) vs. loss
  6. 7. In an image classification problem, an algorithm's ability to successfully classify images even when the orientation of the image changes
  7. 10. A single update of a model's weights during training
  8. 16. A feature not present among the input features, but created from one or more of them
  9. 17. Synonym for loss
  10. 19. A popular clustering algorithm that groups examples in unsupervised learning