awhitefield8 / learning_in_games

Replication code: Competing for No-regret Buyers

Introduction

The repo contains the code that replicates the experiments and figures in 'Competing for No-regret buyers'.

The models.py file contains classes for buyers who play multiplicative weighs.

The funcs.py contains functions for value iteration algorithms that i) approximate best responses to opponent strategies ii) compute equilibria. It also contains some functions for plotting value and policy functions.

Code Structure

├── README.md
├── funcs.py
├── models.py
└── notebooks.ipynb

References

[Learning in Games (And Games in Learning), Aaron Roth]

Authors: Declan Mirabella, Alexander Whitefield