Artificial Intelligence
Vladimir Belik's user avatar
Vladimir Belik's user avatar
Vladimir Belik's user avatar
Vladimir Belik
  • Member for 1 year, 11 months
  • Last seen more than a week ago
Profile Activity
Stats
352
reputation
11k
reached
2
answers
13
questions
Communities
View all
Badges
View all badges
This user doesn’t have any gold badges yet.
3
silver badges
14
bronze badges
Top tags
View all tags
1
Score
10
Posts
67
Posts %
1
Score
5
Posts
33
Posts %
1
Score
2
Posts
13
Posts %
1
Score
2
Posts
13
Posts %
1
Score
1
Posts
7
Posts %
0
Score
8
Posts
53
Posts %
Top posts
View all questions and answers
question
5
Why are only neural networks (and not SVMs, for example) used for reinforcement learning?
Jun 5, 2022
question
5
Why is exploitation necessary during training?
Jun 7, 2022
question
4
Mathematically, what is happening differently in the neural net during exploration vs. exploitation?
Jun 7, 2022
question
3
Does LSTM provide any unique value or advantages compared to other algorithms, including "vanilla" RNN?
Jun 26, 2022
question
2
Why are agents trained in episodes, even in non-episodic tasks?
Jun 8, 2022
question
2
For continuing tasks, is the choice of episode length completely arbitrary?
Jun 5, 2022
question
2
Do I need to normalize all state-space variables? If so, how?
May 29, 2022
question
2
In RL, is it possible to design a multiplicative/exponential reward function? A reward func that depends on current accumulated reward?
May 31, 2022
question
1
How to preserve Markov Property in Deep Reinforcement Learning when using "mixup" or "mixreg"?
Jun 15, 2022
question
1
How to teach DRL agent to avoid erratic, high frequency behavior (when costs for taking actions are already included)?
Jun 26, 2022
Top network posts
View all network posts
16
For neural networks, is mini-batching done purely because of memory constraints?
6
Stable Baselines3 PPO() - how to change clip_range parameter during training?
6
Why is GARCH offering no predictive value?
6
How can I gather data from past cases in the US? Is it possible?
5
Why custom evaluation/scoring metric is causing overfitting in (cross) validation?