DEEP LEARNING CO3

Edit Answers
12345678910
Across
  1. 4. A regularization technique used in neural networks to prevent overfitting by randomly disabling neurons.
  2. 5. Technique used to stabilize learning by normalizing layer inputs during training.
  3. 8. The method of adapting a pre-trained model to a new task.
  4. 10. The first well-known deep learning architecture that won the ImageNet competition in 2012.
Down
  1. 1. that introduce non-linearity into neural networks, such as ReLU or sigmoid.
  2. 2. A model known for its innovative use of multiple convolutional filter sizes within the same layer.
  3. 3. The operation that extracts features from input data by applying filters.
  4. 6. A deep learning architecture known for its use of small convolutional filters and a simple, stacked design.
  5. 7. Architecture that introduced skip connections to mitigate the vanishing gradient problem.
  6. 9. The main elements detected by convolutional layers in a network.