NN-recap

1234567891011
Across
  1. 3. Slope used to update weights during optimization
  2. 4. Learning algorithm that propagates errors backward
  3. 8. Step size when moving filter across input
  4. 11. Downsampling technique that reduces spatial dimensions
Down
  1. 1. Another name for the sliding window in convolution
  2. 2. Regularization technique that randomly ignores neurons during training
  3. 5. Mathematical operation that slides filters across images
  4. 6. Function that introduces non-linearity after weighted sum
  5. 7. Popular activation function that outputs max(0,x)
  6. 9. Complete pass through the entire training dataset
  7. 10. Learnable matrix that detects features in CNNs