Robust Bidding in Learning Classifier Systems Using Loan and Bid History

Two-Step Markov Update Algorithm for Accuracy-Based Learning Classifier Systems

“…Originally, LCS was structured to model a Boolean environment, although numerous studies have been proposed to represent real-valued problems such as XCSR [5,12] and problems with mixed attributes as in AKLR [13]. The robust bidding strategy of strengthbased classifiers is studied in [14], and later in MLCS [15] is extended to handle multi-label data, while the labels are allowed to have confidence levels.…”