A comparison of distributed machine learning methods for the support of "many labs" collaborations in computational modeling of decision making.

Neither CL-based agents (η = 0.005, SE = 0.004, p = 0.26) nor the FL-based and the IL-based agents ( FL-based: η = 0.001, SE = 0.004, p = 0.91, IL-based: η = 0.001, SE = 0.004, p = 0.83) are significantly different for subjects in terms of the probability of selecting DeckA, although the CIL-based selected slightly significant more DeckA than subjects (η = 0.01, SE = 0.005, p = 0.03).

also in 1,762 other papers