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.

Speaker Info:

Linked In Profile

Speaker Links:

Blog

Id: 1654
Section: Data Science, Machine Learning and AI
Type: Poster
Target Audience: Intermediate
Last Updated: