Recessive's user avatar
Recessive's user avatar
Recessive's user avatar
Recessive
  • Member for 4 years, 8 months
  • Last seen more than a month ago
8 votes
2 answers
4k views

Why is gradient descent used over the conjugate gradient method?

8 votes
1 answer
5k views

Why is the learning rate generally beneath 1?

6 votes
2 answers
3k views

Does using the softmax function in Q learning not defeat the purpose of Q learning?

3 votes
0 answers
88 views

Why do regression LSTMs learn high to low inputs significantly better than low to high?

3 votes
1 answer
333 views

In unsupervised learning, what is meant by "finding the probability of an image"?

3 votes
1 answer
440 views

How are exploding numbers in a forward pass of a CNN combated?

2 votes
0 answers
231 views

How is the bias caused by a max pooling layer overcome?

2 votes
1 answer
79 views

How is a neural network where the majority of inputs are 0 trained?

2 votes
1 answer
117 views

Structure discrepancy of an LSTM?

2 votes
1 answer
100 views

Why does an LSTM cycle on initialisation?

2 votes
1 answer
3k views

How does a decision tree split a continuous feature?

1 vote
2 answers
103 views

What are the standard problems for CNNs and LSTMs?

1 vote
1 answer
217 views

Is it possible with stochastic gradient descent for the error to increase?

1 vote
1 answer
267 views

Dropout causes too much noise for network to train

1 vote
1 answer
81 views

Is the following the correct implementation of the Q learning algorithm for a neural network?

1 vote
0 answers
1k views

Is attention always better then an RNN/CNN?

1 vote
0 answers
95 views

When training a seq2seq model is it better to train using the models outputs or expected outputs?

1 vote
0 answers
109 views

How is it possible to use batches of data from within the same sequence with an LSTM?

0 votes
1 answer
141 views

Entirely linear neural network learning non-linear function

0 votes
2 answers
506 views

Possible model to use to find pixel locations of objects

0 votes
1 answer
389 views

How does an LSTM output the correct dimensions for classes?