PropertyValue
nif:beginIndex
  • 0 (xsd:integer)
nif:broaderContext
nif:endIndex
  • 230 (xsd:integer)
nif:isString
  • Essentially, the goal is to minimize the difference between the current estimation of the Q-value (prediction), and an updated estimate (target) that combines the obtained reward and an estimation of the quality of the next state.
rdf:type