Questions tagged [exploration-exploitation-tradeoff]

For questions related to the exploration-exploitation trade-off (or dilemma) in the context of reinforcement learning.

Filter by
Sorted by
Tagged with
7
votes
1answer
227 views

What is the purpose of the actor in actor-critic algorithms?

For discrete action spaces, what is the purpose of the actor in actor-critic algorithms? My current understanding is that the critic estimates the future reward given an action, so why not just take ...
5
votes
3answers
462 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 ...
4
votes
2answers
244 views

What is the meaning of “exploration” in reinforcement and supervised learning?

While exploration is an integral part of reinforcement learning (RL), it does not pertain to supervised learning (SL) since the latter is already provided with the data set from the start. That said, ...
3
votes
1answer
60 views

Can we stop training as soon as epsilon is small?

I'm new to reinforcement learning. As it is common in RL, $\epsilon$-greedy search for the behavior/exploration is used. So, at the beginning of the training, $\epsilon$ is high, and therefore a lot ...
3
votes
2answers
103 views

Why is it not advisable to have a 100 percent exploration rate? [duplicate]

During the learning phase, why don't we have a 100% exploration rate, to allow our agent to fully explore our environment and update the Q values, then during testing we bring in exploitation? Does ...
2
votes
1answer
75 views

Why is 100% exploration bad during the learning stage in reinforcement learning?

Why can't we during the first 1000 episodes allow our agent to perform only exploration? This will give a better chance of covering the entire space state. Then, after the number of episodes, we can ...
2
votes
1answer
256 views

Does eligibility traces and epsilon-greedy do the same task in different ways?

I understand that, in Reinforcement Learning algorithms, such as Q-learning, to prevent selecting the actions with greatest q-values too fast and allow for exploration, we use eligibility traces. Here ...
2
votes
1answer
54 views

Why do some DQN implementations not require random exploration but instead emulate all actions?

I've found online some DQN algorithms that (in a problem with a continuous state space and few actions, let's say 2 or 3), at each time step, compute and store (in the memory used for updating) all ...
1
vote
2answers
75 views

Why can't we fully exploit the environment after the first episode in Q-learning?

During the first episode, it's 100% exploration, because all our Q values are 0. Suppose we have 1000 time steps, and it's terminated by meeting a reward. So, after the first episode, why can't we ...
1
vote
2answers
107 views

Should I use exploration strategy in Policy Gradient algorithms?

In policy gradient algorithms the output is a stochastic policy - a probability for each action. I believe that if I follow the policy (sample an action from the policy) I make use of exploration ...
1
vote
1answer
51 views

What is the optimal exploration-exploitation trade-off in Q*bert?

I am training an RL agent with Deep Q-learning + Experience Replay on the Q*bert Atari environment. After 400,000 frames, my agent appears to have learned strategic information about the game, but ...
1
vote
1answer
105 views

Can tabular Q-learning converge even if it doesn't explore all state-action pairs?

My understanding of tabular Q-learning is that it essentially builds a dictionary of state-action pairs, so as to maximize the Markovian (i.e., step-wise, history-agnostic?) reward. This incremental ...
1
vote
1answer
37 views

Should I just use exploitation after I have trained the Q agent?

When using a trained Q-learning algorithm in an actual game, would I just use exploitation and no longer use exploration? Should I use exploration only during the training phase?
1
vote
0answers
39 views

Why is the ideal exploration parameter in the UCT algorithm $\sqrt{2}$?

From Wikipedia, in the Monte-Carlo Tree Search algorithm, you should choose the node that maximizes the value: $${\displaystyle {\frac {w_{i}}{n_{i}}}+c{\sqrt {\frac {\ln N_{i}}{n_{i}}}}},$$ where ${...
1
vote
0answers
46 views

What is the difference between exploitation and exploration in the context of optimization?

In the paper Moth-flame optimization algorithm: A novel nature-inspired heuristic paradigm (2015, published in Knowledge-Based Systems) The test functions are divided to three groups: unimodal, multi-...
1
vote
0answers
28 views

Should the exploration rate be updated at the end of the episode or at every step?

My agent uses an $\epsilon$-greedy strategy to learn. The exploration rate (i.e. $\epsilon$) decays throughout the training. I've seen examples where people update $\epsilon$ every time an action is ...
1
vote
1answer
38 views

How to deal with the addition of a new state to the environment during training?

Let's say we have a dynamic environment: a new state gets added after 2000 episodes have been done. So, we leave room for exploration, so that it can discover the new state. When it gets to that new ...
1
vote
2answers
57 views

Why do we explore after we have an accurate estimate of the value function?

Suppose we have a small space state and that, after about 2000 episodes, we've accurately explored the environment and known the accurate $Q$ values. In that case, why do we still leave a small ...
1
vote
0answers
35 views

How can I increase the exploration in the Proximal Policy Optimation algorithm?

How can I increase the exploration in the Proximal Policy Optimation reinforcement learning algorithm? Is there a variable assigned for this purpose? I'm using the stable-baseline implementation: ...
1
vote
0answers
292 views

Why is the $\epsilon$ hyper-parameter (in the $\epsilon$-greedy policy) annealed smoothly?

As far as I understand, RL is a process that can be divided into 2 stages: Exploring a wide range of paths (acting randomly) Refining the current optimal paths (revolving around actions with a so-...
0
votes
0answers
22 views

PPO: sampling next action vs picking the most probable action

According to the original Proximal Policy Optimization paper (PPO paper), we always sample an action from the actor distribution. According to the link The overall loss is calculated as $\text{loss} =...
0
votes
0answers
56 views

How to modify the Actor-Critic policy gradient algorithm to perform Safe exploration in Reinforcement Learning

I am trying to implement safe exploration technique in [Ref.1]. I am using Soft Actor-Critic algorithm to teach an agent to introduce a bias between 0 and 1 to a specific state of interest in my ...