Two share links about bandit algorithm

最新推荐文章于 2024-12-13 01:49:22 发布

fibo_nacci

最新推荐文章于 2024-12-13 01:49:22 发布

阅读量632

点赞数

CC 4.0 BY-SA版权

分类专栏： ML

本文链接：https://blog.youkuaiyun.com/flyupliu/article/details/24392629

ML 专栏收录该内容

1 篇文章

订阅专栏

博客分享了多臂老虎机算法相关内容，介绍了文章发布链接，还提出什么是多臂老虎机算法，指出一个臂代表一个选项。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

此文同时在 http://kunth.github.io/2014/04/23/bandit-algorithm.html

Two share links about bandit algorithm

what's (multi-arm) bandit algorithm ?

epsilon-greedy algorithm

With probability 1 – epsilon, the epsilon-Greedy algorithm exploits the best known option.
• With probability epsilon / 2, the epsilon-Greedy algorithm explores the best known option.
• With probability epsilon / 2, the epsilon-Greedy algorithm explores the worst known option.

one arm denotes one option

when pulled, any given arm will output a reward.
you need to cope with risk by figuring out which arm has the highest average reward.
what makes a bandit problem special is that we only receive a small amount of the information about the rewards from each arm.
we only find out about the reward that was given out by the arm we actually pulled.
whichever arm we pull, we miss out on information about hte other arms that we didn't pull.
every time we experiment with an arm that isn't the best arm, we lose reawrd because we could, at least in principle, have pulled on a better arm.

Here are two links about multi-bandit algorithm that may help you

Bndit algorithm for website optimization
Algorithms for the multi-armed bandit problem