Human decision making is error-prone and often subject to biases. Important information cues are misweighted and feedback delays hamper learning. Experimentally, task information ...
Inthispaper,weproposeadescriptorcombination method, which enables to improve significantly the recognition rate compared to the recognition rates obtained by each descriptor. This ...
Improvisation is rapidly becoming an important issue for both scholars and practitioners. Organizations that operate in turbulent environments must learn to swiftly adapt and resp...
This paper proposes and experimentally validates a Bayesian network model of a range finder adapted to dynamic environments. All modeling assumptions are rigorously explained, and...
Tinne De Laet, Joris De Schutter, Herman Bruyninck...
Abstract. Recently, on-line adaptation of binary classifiers for tracking have been investigated. On-line learning allows for simple classifiers since only the current view of the ...