http://popups.lib.uliege.be/1373-5411/index.php?id=2258 Publications of Auteurs Daisuke Uragami fr 0 http://popups.lib.uliege.be/1373-5411/index.php?id=3065 In incremental learning, it is necessary to conquer the dilemma of plasticity and stability. Because neural networks usually employ continuously distributed representation for state space, learning newly added data affects the existing memories. We apply a neural network with algebraic (lattice) structure to incremental learning, that has been proposed to model information processing in the dendrites of neurons. It has been proposed as a mathematical model of information processing in the dendrites of neurons. Because of the operation 'maximum' in lattice algebra weakening the continuously distributed representation, our proposed model succeeds in incremental learning. Fri, 06 Sep 2024 16:05:50 +0200 Fri, 06 Sep 2024 16:06:03 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=3065 http://popups.lib.uliege.be/1373-5411/index.php?id=2253 It has been recognized that there are two kinds of cognitive maps, route map and survey map. Researchers recently have focused on the integration of both maps. The attempt to integrate them, however, entails a problem ; which type is prior to the other. This reveals an infinite regression. We propose that the paradoxical modality resulting from the integration can be expressed as a heterarchy, a dynamical hierarchical system. In the name of heterarchy, we stress negotiation among levels. If there is a discrepancy between levels, the expression of a level and interaction are destined to contain intrinsic indefiniteness. It reveals the negotiation. Route maps and survey maps are not integrated but negotiated, and those form a heterarchy. We conducted a particular experiment in which negotiations between route maps and survey maps were enhanced. Tue, 30 Jul 2024 15:41:31 +0200 Tue, 30 Jul 2024 15:41:39 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=2253