Update.Evaluator.QLearningPersExpPath: Controlled Copying of Models
In MartinKies/RLR: Reinforcement Learning with R

Returns a new, updated evaluator

Update.Evaluator.QLearningPersExpPath(
  evaluator = evaluator,
  new.model = NULL,
  new.model.surp = NULL,
  new.model.fam = NULL,
  which.update = "both",
  model.par
)

`evaluator`	Evaluation consisting of the two models
`new.model`	Model with which we want to update. If `NULL`, no update is done.
`new.model.surp`	If not `NULL`, the surprise model is updated accordingly.
`new.model.fam`	If not `NULL`, the familiarity model is updated accordingly.
`which`	May be 'both', 'cur' or 'best' and updates respectively.