Gulzar's user avatar
Gulzar's user avatar
Gulzar's user avatar
Gulzar
  • Member for 5 years, 3 months
  • Last seen more than a month ago
14 votes
3 answers
13k views

Why does is make sense to normalize rewards per episode in reinforcement learning?

6 votes
1 answer
3k views

Can TD($\lambda$) be used with deep reinforcement learning?

4 votes
1 answer
958 views

Meaning of Actor Output in Actor Critic Reinforcement Learning

4 votes
2 answers
404 views

How to identify too small network in reinforcement learning?

4 votes
1 answer
562 views

Difference between dot product attention and "matrix attention"

3 votes
1 answer
500 views

What information should be cached in experience replay for actor-critic?

3 votes
2 answers
340 views

How to obtain a formula for loss, when given an iterative update rule in gradient descent?

2 votes
1 answer
1k views

In online one step actor critic, why does the weights update become less significant as the episode progresses?

2 votes
1 answer
1k views

Why is the reward signal normalized in openAI's REINFORCE? [duplicate]

2 votes
1 answer
577 views

Which Openai Gym environment should I use to test a Temporal Difference RL Agent?

2 votes
1 answer
4k views

What is a cascaded convolutional neural network?

2 votes
1 answer
125 views

Is there a relation between the size of the neural networks and speed of convergence in deep reinforcement learning?

2 votes
2 answers
961 views

Why does triplet loss allow to learn a ranking whereas contrastive loss only allows to learn similarity?

1 vote
0 answers
323 views

variational auto encoder loss goes down but does not reconstruct input. out of debugging ideas

1 vote
0 answers
92 views

Underfitting a single batch: Can't cause autoencoder to overfit multi-sample batches of 1d data. How to debug?

1 vote
0 answers
29 views

Dealing with bias in multi-channel auto encoders

1 vote
1 answer
166 views

How to define a "don't care" class in time series classification in Pytorch?

1 vote
1 answer
105 views

How to train an LSTM to classify based on rare historic event?

1 vote
2 answers
156 views

When are Q values calculated in experience replay?

1 vote
1 answer
44 views

Trained a regression network and getting EXACT same result on validation set, on every epoch

1 vote
0 answers
133 views

How to make episode ending "good" in reinforcement learning?

0 votes
0 answers
35 views

Correct way to work with both categorical and continuous features together

0 votes
1 answer
1k views

Why is this PyTorch implementation of the actor-critic algorithm inconsistent with the mathematical formulas?

0 votes
1 answer
47 views

How to (theorically) build a neural network with input of size 0?

0 votes
1 answer
411 views

What does the Critic network evaluate in Actor Critic?

0 votes
0 answers
2k views

Focal Loss vs Weighted Cross Entropy Loss

0 votes
0 answers
29 views

What causes train loss to suddenly rise, and never fall back, starting at some epoch?

-1 votes
1 answer
182 views

Understanding self attention - How come there is no connection between different states?