medium10 pts · first answer countsdeep-learningactivations
Neurons that always output zero
After training with a large learning rate, 40% of the ReLU units in a layer output 0 for every input. What is this and a common remedy?
Discussion 0
Join free to post in the discussion.
Related questions
- ReLU applied to a vectorMachine Learning Engineer
- Parameters in nn.Linear(10, 5)Machine Learning Engineer
- Why sigmoid slows deep trainingMachine Learning Engineer
- Range of tanhMachine Learning Engineer
- Cross-entropy loss by handMachine Learning Engineer
- Telling autograd to track a tensorMachine Learning Engineer
- Why wrap inference in no_gradMachine Learning Engineer
- Logging loss without leaking memoryMachine Learning Engineer