Ray Walker's user avatar
Ray Walker's user avatar
Ray Walker's user avatar
Ray Walker
  • Member for 2 years, 6 months
  • Last seen more than 1 year ago
40 votes
2 answers
18k views

What is the relation between Q-learning and policy gradients methods?

  • 525
13 votes
3 answers
5k views

Why exactly do neural networks require i.i.d. data?

  • 35k
12 votes
1 answer
5k views

What is the relation between online (or offline) learning and on-policy (or off-policy) algorithms?

  • 35k
10 votes
1 answer
2k views

Do off-policy policy gradient methods exist?

  • 653
10 votes
1 answer
3k views

What is the difference between expected return and value function?

8 votes
2 answers
1k views

Why are lambda returns so rarely used in policy gradients?

7 votes
1 answer
498 views

Why is the state-action value function used more than the state value function?

7 votes
2 answers
763 views

Why are reinforcement learning methods sample inefficient?

  • 253
6 votes
1 answer
352 views

What is the relation between a policy which is the solution to a MDP and a policy like $\epsilon$-greedy?

  • 35k
5 votes
1 answer
572 views

Why not more TD(𝜆) in actor-critic algorithms?

  • 125
5 votes
2 answers
330 views

Is it possible to guide a reinforcement learning algorithm?

5 votes
2 answers
2k views

How can I handle overfitting in reinforcement learning problems?

  • 2,569
4 votes
1 answer
266 views

Immediate reward received in Atari game using DQN

  • 1,241
4 votes
1 answer
1k views

Why are model-based methods more sample efficient than model-free methods?

  • 431
3 votes
1 answer
744 views

How does the initialization of the value function and definition of the reward function affect the performance of the RL agent?

3 votes
0 answers
125 views

Can I use deterministic policy gradient methods for stochastic policy learning?

3 votes
2 answers
655 views

Apart from the state and state-action value functions, what are other examples of value functions used in RL?

  • 35k
3 votes
1 answer
54 views

Can policy iteration use only the immediate reward for updates?

2 votes
3 answers
102 views

How to estimate a behavior policy for off-policy learning based on data?

1 vote
0 answers
59 views

Why is the number of neurons used in various neural networks power of 2?

  • 126