Edamame's user avatar
Edamame's user avatar
Edamame's user avatar
Edamame
  • Member for 6 years, 6 months
  • Last seen more than 1 year ago
  • San Francisco Bay Area
181 votes
5 answers
129k views

What is the "dying ReLU" problem in neural networks?

  • 3,845
86 votes
7 answers
61k views

Time series prediction using ARIMA vs LSTM

  • 1,075
71 votes
6 answers
139k views

Cross-entropy loss explanation

  • 2,971
65 votes
4 answers
69k views

Why mini batch size is better than one single "batch" with all training data?

  • 8,047
62 votes
4 answers
75k views

Does batch_size in Keras have any effects in results' quality?

  • 1,140
36 votes
4 answers
40k views

What is the meaning of "The number of units in the LSTM cell"?

  • 829
14 votes
2 answers
7k views

What is the difference between an autoencoder and an encoder-decoder?

  • 594
10 votes
1 answer
7k views

Q-Learning: Target Network vs Double DQN

  • 2,586
9 votes
1 answer
4k views

Where does the name 'LSTM' come from?

  • 17.9k
5 votes
2 answers
5k views

What algorithms will stuck in the local minimum?

  • 2,525
4 votes
1 answer
214 views

How to deal with annotation errors?

  • 2,525
3 votes
1 answer
783 views

What are the output shape of LSTM with "return_sequences" equal to "True" or "False"?

  • 1,152
3 votes
1 answer
107 views

reinforcement learning: Decompose a policy gradient

  • 2,525
2 votes
1 answer
328 views

Training Encoder-Decoder using Decoder Outputs

  • 207
2 votes
1 answer
1k views

LSTM input in Keras

  • 23
1 vote
2 answers
103 views

Intuition behind the entropy definition

  • 2,525
1 vote
1 answer
1k views

Why would batch normalization allows us to use higher learning rate in the neural network?

  • 2,525