Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Bayes vs Reinforcement

No description

Gordon Ovens

on 24 February 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Bayes vs Reinforcement

Bayesian updating
Reinforcement Heuristic Bayesian Updating (BEU) Using prior learned information when making decisions according to the laws of probability Reinforcement Heuristic (RH) - Win stay/ lose shift
- Pick choices associated with successful past outcomes The Experiment - First draw right BEU and RH coincide
- First draw left BEU and RH oppose - Treatment 1 & 2 same method, paid for drawing black balls
- Each treatment 60 periods of 2 draws
Main Findings - After right draw BEU and RH consistent (very low error rates)
- After left draw about half of participants (47.5% treatment 1 and 48.3% treatment 2) use RH and the rest BEU
- Treatment 3 only 28.2% use RH so removing the affect of the first draw reduces error rates People are different with about 50% following Bayesian principles and about 50% using a simplifying heuristic Take home message - Treatments 3 first draw was purely informational to try to avoid any form of affect
- 80 periods of 2 draws Dan Levin Gary Charness
Full transcript