The Art of Choosing Blindly : Multi-Armed Bandits
Aniq Ur Rahman (~Aniq55) |
Description:
- Brief description of Multi-armed Bandits (MAB)
- Areas of applying the MAB framework
- Connection to Reinforcement Learning and various algorithms
- A full-fledged Python Library for implementing the MAB framework in versatile problems
- Sample Python scripts for implementing some algorithms like Thompson Sampling, UCB etc.
Prerequisites:
- Python OOP
- Love for Python
Content URLs:
-
Poster [ draft ]
-
A work based on Multi-Armed Bandit was presented at WMLC - WiOpt '19 Avignon, France. [ Available here ]
-
Github repository will be made public post-event.