Education
TeachingviaBest-CaseCounterexamples intheLearning-with-Equivalence-QueriesParadigm
We establish new connections between LwEQ-TD and LfS-TD by studying LwEQ-TD for different learner models based on the richness of their query functions. We show that LwEQ-TD isthesameaswc-TD[18],RTD[22,24],andNCTD[27]forahypothesis class when restricting query functions to specific families.
RUDDER: Return Decomposition for Delayed Rewards
Jose A. Arjona-Medina, Michael Gillhofer, Michael Widrich, Thomas Unterthiner, Johannes Brandstetter, Sepp Hochreiter
Neural Information Processing Systems http://nips.cc/
Universal Ratesfor Interactive Learning
Definition 2(Informal (see Definition 5)). Definition 3(Informal (see Definition 6)).
NearlyOptimalBest-of-Both-WorldsAlgorithmsfor OnlineLearningwithFeedbackGraphs
This study considers online learning with general directed feedback graphs.
NearlyOptimalBest-of-Both-WorldsAlgorithmsfor OnlineLearningwithFeedbackGraphs
This study considers online learning with general directed feedback graphs.