Skip to contents

Temperature functions

Corson N. Areshenkoff

2024-08-13

temperature_functions.Rmd

iowa implements multiple common temperature functions, and can be relatively easily extended to include more. These functions control the temperature of the softmax function use to assign probabilities to decks. Currently, The following are included in the base package:

Time independent parameters

  • Keyword: TIC
  • Parameters: c (Temperature)
  • Bounds: \(c \in [0,\infty)\)

\[ \theta = 3^c \]

Time dependent parameters

  • Keyword: TDC
  • Parameters: c (Temperature)
  • Bounds: \(c \in [0,\infty)\)
  • Description: Temperature increases with increasing trials. Intended to simulate a gradual shift from exploratory to exploitative behavior.

\[ \theta(t) = \left ( \frac{t}{10} \right )^c \]