Questions tagged [q-learning]

For questions related to the Q-learning algorithm, which is a model-free and temporal-difference reinforcement learning algorithm that attempts to approximate the Q function, which is a function that, given a state s and an action a, returns a real number that represents the return (or value) of state s when action a is taken from s. Q-learning was introduced in the PhD thesis "Learning from Delayed Rewards" (1989) by Watkins.

Filter by
Sorted by
Tagged with
32
votes
2answers
12k views

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

As far as I understand, Q-learning and policy gradients (PG) are the two major approaches used to solve RL problems. While Q-learning aims to predict the reward of a certain action taken in a certain ...
17
votes
3answers
2k views

Why doesn't Q-learning converge when using function approximation?

The tabular Q-learning algorithm is guaranteed to find the optimal $Q$ function, $Q^*$, provided the following conditions (the Robbins-Monro conditions) regarding the learning rate are satisfied $\...
14
votes
1answer
3k views

Why does DQN require two different networks?

I was going through this implementation of DQN and I see that on line 124 and 125 two different Q networks have been initialized. From my understanding, I think one network predicts the appropriate ...
11
votes
2answers
1k views

How do we prove the n-step return error reduction property?

In section 7.1 (about the n-step bootstrapping) of the book Reinforcement Learning: An Introduction (2nd edition), by Andrew Barto and Richard S. Sutton, the authors write about what they call the "n-...
11
votes
2answers
4k views

Can Q-learning be used for continuous (state or action) spaces?

Many examples work with a table-based method for Q-learning. This may be suitable for a discrete state (observation) or action space, like a robot in a grid world, but is there a way to use Q-learning ...
10
votes
1answer
579 views

Are Q-learning and SARSA the same when action selection is greedy?

I'm currently studying reinforcement learning and I'm having difficulties with question 6.12 in Sutton and Barto's book. Suppose action selection is greedy. Is Q-learning then exactly the same ...
9
votes
2answers
1k views

What is the difference between Q-learning, Deep Q-learning and Deep Q-network?

Q-learning uses a table to store all state-action pairs. Q-learning is a model-free RL algorithm, so how could there be the one called Deep Q-learning, as deep means using DNN; or maybe the state-...
8
votes
1answer
609 views

Does AlphaZero use Q-Learning?

I was reading the AlphaZero paper Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm, and it seems they don't mention Q-Learning anywhere. So does AZ use Q-...
8
votes
1answer
2k views

What are other ways of handling invalid actions in scenarios where all rewards are either 0 (best reward) or negative?

I created an OpenAI Gym environment, and I would like to check the performance of the agent from OpenAI Baselines DQN approach on it. In my environment, the best possible outcome for the agent is 0 - ...
7
votes
2answers
730 views

What does the symbol $\mathbb E$ mean in these equations?

I came across some papers that use $\mathbb E$ in equations, in particular, this paper: https://arxiv.org/pdf/1511.06581.pdf. Here is some equations from the paper that uses it: $Q^\pi \left(s,a \...
6
votes
2answers
1k views

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

It is my understanding that, in Q-learning, you are trying to mimic the optimal $Q$ function $Q*$, where $Q*$ is a measure of the predicted reward received from taking action $a$ at state $s$ so that ...
6
votes
1answer
2k views

Can Q-learning be used in a POMDP?

Can Q-learning (and SARSA) be directly used in a Partially Observable Markov Decision Process (POMDP)? If not, why not? My intuition is that the policies learned will be terrible because of partial ...
6
votes
1answer
204 views

Q-Learning the generic maze solution

After doing some exercices on Q-learning for maze solving, I wondered : my q-learning algorithms solve only ONE maze. The AI doesn't learn how to solve mazes, so how can I achieve it ? For instance ...
6
votes
1answer
2k views

How does Q-learning work in stochastic environments?

The Q function uses the (current and future) states to determine the action that gets the highest reward. However, in a stochastic environment, the current action (at the current state) does not ...
6
votes
1answer
3k views

How to implement exploration function and learning rate in Q Learning

I'm trying to implement Q-learning (state-based representation and no neural / deep stuff) but I'm having a hard time getting it to learn anything. I believe my issue is with the exploration function ...
6
votes
1answer
149 views

How is the DQN loss derived from (or theoretically motivated by) the Bellman equation, and how is it related to the Q-learning update?

I'm doing a project on Reinforcement Learning. I programmed an agent that uses DDQN. There are a lot of tutorials on that, so the code implementation was not that hard. However, I have problems ...
6
votes
2answers
584 views

Can DQN perform better than Double DQN?

I'm training both DQN and double DQN in the same environment, but DQN performs significantly better than double DQN. As I've seen in the double DQN paper, double DQN should perform better than DQN. Am ...
6
votes
0answers
164 views

Is the Bellman equation that uses sampling weighted by the Q values (instead of max) a contraction?

It is proved that the Bellman update is a contraction (1). Here is the Bellman update that is used for Q-Learning: $$Q_{t+1}(s, a) = Q_{t}(s, a) + \alpha*(r(s, a, s') + \gamma \max_{a^*} (Q_{t}(s', ...
5
votes
3answers
1k views

What is the target Q-value in DQNs?

I understand that in DQNs, the loss is measured by taking the MSE of outputted Q-values and target Q-values. What does the target Q-values represent? And how is it obtained/calculated by the DQN?
5
votes
2answers
1k views

Is Q-learning a type of model-based RL?

Model-based RL creates a model of the transition function. Tabular Q-Learning does this iteratively (without directly optimizing for the transition function). So, does this make tabular Q-learning a ...
5
votes
2answers
297 views

What are some online courses for deep reinforcement learning?

What are some (good) online courses for deep reinforcement learning? I would like the course to be both programming and theoretical. I really liked David Silver's course, but the course dates from ...
5
votes
1answer
500 views

Is the discount not needed in a deterministic environment for Reinforcement Learning?

I'm now reading a book titled as "Deep Reinforcement Learning Hands-On" and the author said the following on the chapter about AlphaGo Zero: Self-play In AlphaGo Zero, the NN is used to ...
5
votes
3answers
1k views

What is a high dimensional state in reinforcement learning?

In the DQN paper, it is written that the state-space is high dimensional. I am a little bit confused about this terminology. Suppose my state is a high dimensional vector of length $N$, where $N$ is a ...
5
votes
1answer
2k views

How should I model all available actions of a chess game in deep Q-learning?

I just read about deep Q-learning, which is using a neural network for the value function instead of a table. I saw the example here: Using Keras and Deep Q-Network to Play FlappyBird and he used a ...
5
votes
1answer
180 views

Why does Q-learning converge under 100% exploration rate?

I am working on this assignment where I made the agent learn state-action values (Q-values) with Q-learning and 100% exploration rate. The environment is the classic gridworld as shown in the ...
5
votes
3answers
463 views

In Q-learning, wouldn't it be better to simply iterate through all possible states?

In Q-learning, all resources I've found seem to say that the algorithm to update the Q-table should start at some initial state, and pick actions (which are sometimes random) to explore the state ...
5
votes
1answer
80 views

How would I compute the optimal state-action value for a certain state and action?

I am currently trying to learn reinforcement learning and I started with the basic gridworld application. I tried Q-learning with the following parameters: Learning rate = 0.1 Discount factor = 0.95 ...
5
votes
1answer
635 views

What are good learning strategies for Deep Q-Network with opponents?

I am trying to find out what are some good learning strategies for Deep Q-Network with opponents. Let's consider the well-known game Tic-Tac-Toe as an example: How should an opponent be implemented ...
5
votes
1answer
123 views

Can exogenous variables be state features in reinforcement learning?

I have a question about state representation of Q-learning or DQN algorithm. I'm still a beginner of RL, so I'm not sure that is it suitable to take exogenous variables as state features. For example,...
5
votes
1answer
132 views

Reinforcement Learning (Fitted Q): Qn on Concept & Implementation

I hope to get some clarifications on Fitted Q-Learning ('FQL'). My Research So Far I've read Sutton's book (specifically, chp 6 to 10), Ernst et al and this paper. I know that ...
5
votes
1answer
1k views

Why is the target $r + \gamma \max_{a'} Q(s', a'; \theta_i^-)$ in the loss function of the DQN architecture?

In the paper Human-level control through deep reinforcement learning, the DQN architecture is presented, where the loss function is as follows $$ L_i(\theta_i) = \mathbb{E}_{(s, a, r, s') \sim U(D)} \...
5
votes
4answers
725 views

How to stop DQN Q function from increasing during learning?

Following the DQN algorithm with experience replay: Store transition $\left(\phi_{t}, a_{t}, r_{t}, \phi_{t+1}\right)$ in $D$ Sample random minibatch of transitions $\left(\phi_{j}, a_{j}, r_{j}, \...
4
votes
2answers
629 views

Is there any good reference for double deep Q-learning?

I am new in reinforcement learning, but I already know deep Q-learning and Q-learning. Now, I want to learn about double deep Q-learning. Do you know any good references for double deep Q-learning? ...
4
votes
3answers
451 views

Upper limit to the maximum cumulative reward in a deep reinforcement learning problem

Is there an upper limit to the maximum cumulative reward in a deep reinforcement learning problem? For example you want to train an DQN agent in an environment and you want to know what is the highest ...
4
votes
1answer
271 views

When do SARSA and Q-Learning converge to optimal Q values?

Here's another interesting multiple-choice question that puzzles me a bit. In tabular MDPs, if using a decision policy that visits all states an infinite number of times, and in each state, randomly ...
4
votes
1answer
138 views

Are Q values estimated from a DQN different from a duelling DQN with the same number of layers and filters?

I am confused about the Q values of a duelling deep Q network (DQN). As far as I know, duelling DQNs have 2 outputs Advantage: how good it is to be in a particular state $s$ Value: the advantage of ...
4
votes
1answer
81 views

Deep Q-Learning “catastrophic drop” reasons?

I am implementing some "classical" papers in Model Free RL like DQN, Double DQN, and Double DQN with Prioritized Replay. Through the various models im running on ...
4
votes
1answer
238 views

Why do DQNs tend to forget?

Why do DQNs tend to forget? Is it because when you feed highly correlated samples, your model (function approximation) doesn't give a general solution? For example: I use level 1 experiences, my ...
4
votes
1answer
209 views

Why can't we apply value iteration when we do not know the reward and transition functions, and how does Q-learning solve this issue?

I don't understand why we can't apply value iteration when don't know the reward and transition probabilities. In this lecture, the lecturer says it has to do with not being able to take max with ...
4
votes
1answer
431 views

State representation of position in 2D plane for Reinforcement Learning (Q Learning)

I recently finished Course on RL by David Silver (on YT) and thought about trying it out on simple application in Unity Game Engine, where I've built simple labyrint with ball and want to teach the ...
4
votes
1answer
82 views

How should I handle invalid actions in a grid world?

I'm building a really simple experiment, where I let an agent move from the bottom-left corner to the upper-right corner of a $3 \times 3$ grid world. I plan to use DQN to do this. I'm having trouble ...
4
votes
1answer
185 views

Is tabular Q-learning considered interpretable?

I am working on a research project in a domain where other related works have always resorted to deep Q-learning. The motivation of my research stems from the fact that the domain has an inherent ...
4
votes
2answers
205 views

What is the difference between return and expected return?

At a time step $t$, for a state $S_{t}$, the return is defined as the discounted cumulative reward from that time step $t$. If an agent is following a policy (which in itself is a probability ...
4
votes
1answer
111 views

Q-learning, am I interpreting correctly $Q(s,a) = r + \gamma \max_{a'} Q(s',a')$?

Ok, due to previous question I was pointed to use reinfrocement learning. So far what I understood from random websites is the following: there is a Q(s,a) function involved I can assume my neural ...
4
votes
2answers
309 views

Should the actor or actor-target model be used to make predictions after training is complete (DDPG)?

The situation I am referring to the paper T. P. Lillicrap et al, "Continuous control with deep reinforcement learning" where they discuss deep learning in the context of continuous action spaces ("...
4
votes
1answer
328 views

How can a DQN backpropagate its loss?

I'm currently trying to take the next step in deep learning. I managed so far to write my own basic feed-forward network in python without any frameworks (just numpy and pandas), so I think I ...
4
votes
1answer
69 views

How to apply or extend the $Q(\lambda)$ algorithm to semi-MDPs?

I want to model an SMDP such that time is discretized and the transition time between the two states follows an exponential distribution and there would be no reward between the transition. Can I ...
4
votes
1answer
113 views

Research into social behavior in Prisoner's Dilemma

I've been working on research into reproducing social behavior using multi-agent reinforcement learning. My focus has been on a GridWorld-style game, but I was thinking that maybe a simpler Prisoner's ...
4
votes
0answers
75 views

Why is there inconsistency in the definitions of the retrace?

In Section 4.3 of paper Learning by Playing - Solving Sparse Reward Tasks from Scratch, the authors define Retrace as $$ Q^{ret}=\sum_{j=i}^\infty\left(\gamma^{j-i}\prod_{k=i}^jc_k\right)[r(s_j,a_j)+\...
4
votes
1answer
130 views

Why Q2 is a more or less independant estimate in Twin Delayed DDPG (TD3)?

Twin Delayed Deep Deterministic (TD3) policy gradient is inspired by both double Q-learning and double DQN. In double Q-learning, I understand that Q1 and Q2 are independent because they are trained ...

1
2 3 4 5
7