PropertyValue
nif:beginIndex
  • 0 (xsd:integer)
nif:broaderContext
nif:endIndex
  • 239 (xsd:integer)
nif:isString
  • (2) When the state-action space is small enough for the Q-values to be represented as a lookup table, this iterative approximation is proved to converge to the true Q-values [15], provided that all state-action pairs are regularly sampled.
rdf:type