engine_RNN: The Engine of Recurrent Neural Network (RNN)

In multiRL: Reinforcement Learning Tools for Multi-Armed Bandit

The Engine of Recurrent Neural Network (RNN)

Description

Because TensorFlow requires numeric arrays and input parameters to learn the mapping between them when building a Recurrent Neural Network (RNN) model, this function transforms simulated data into a standardized dataset and invokes TensorFlow to train the model.

Because TensorFlow requires numeric arrays and input parameters to learn the mapping between them when building a Recurrent Neural Network (RNN) model, this function transforms simulated data into a standardized dataset and invokes TensorFlow to train the model.

Usage

engine_RNN(
  data,
  colnames,
  behrule,
  model,
  funcs = NULL,
  priors,
  settings = NULL,
  control = control,
  ...
)

engine_RNN3(
  data,
  colnames,
  behrule,
  model,
  funcs = NULL,
  priors,
  settings = NULL,
  control = control,
  ...
)

Arguments

data	A data frame in which each row represents a single trial, see data
colnames	Column names in the data frame, see colnames
behrule	The agent's implicitly formed internal rule, see behrule
model	Reinforcement Learning Model
funcs	The functions forming the reinforcement learning model, see funcs
priors	Prior probability density function of the free parameters, see priors
settings	Other model settings, see settings
control	Settings manage various aspects of the iterative process, see control
...	Additional arguments passed to internal functions.

Value

A specialized Recurrent Neural Network (RNN) object. The model can be used with the predict() function to make predictions on a new data frame, estimating the input parameters that are most likely to have generated the given dataset.

A specialized Recurrent Neural Network (RNN) object. The model can be used with the predict() function to make predictions on a new data frame, estimating the input parameters that are most likely to have generated the given dataset.

multiRL documentation built on March 31, 2026, 5:06 p.m.