https://github.com/jldbc/bandits

Multi-Armed Bandit algorithms applied to the MovieLens 20M dataset
https://github.com/jldbc/bandits

Last synced: 5 days ago
JSON representation

Multi-Armed Bandit algorithms applied to the MovieLens 20M dataset

• Host: GitHub
• URL: https://github.com/jldbc/bandits
• Owner: jldbc
• License: mit
• Created: 2019-11-29T21:14:14.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2020-08-09T05:57:24.000Z (over 4 years ago)
• Last Synced: 2025-03-31T13:26:03.091Z (24 days ago)
• Language: Python
• Homepage:
• Size: 4.82 MB
• Stars: 56
• Watchers: 2
• Forks: 17
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Multi-Armed Bandits

Implementations of UCB1, Bayesian UCB, Epsilon Greedy, and EXP3 bandit algorithms on the Movielens-20m dataset. Algorithms are evaluated offline using replay. 

To reproduce: 

```

git clone https://github.com/jldbc/bandits

cd bandits/bandits

bash run.sh

```

[Experiment setup details](https://jamesrledoux.com/algorithms/offline-bandit-evaluation/)

[Impementation details and results](https://jamesrledoux.com/algorithms/bandit-algorithms-epsilon-ucb-exp-python/)

Final results: