no code implementations • 17 Aug 2021 • Leonardo Enzo Brito da Silva, Nagasharath Rayapati, Donald C. Wunsch II
The model achieves improved accuracy and robustness to ordering effects by integrating an online iCVI framework as module B of a topological adaptive resonance theory predictive mapping (TopoARTMAP) -- thereby being named iCVI-TopoARTMAP -- and by employing iCVI-driven post-processing heuristics at the end of each learning step.
1 code implementation • 22 Aug 2020 • Leonardo Enzo Brito da Silva, Nagasharath Rayapati, Donald C. Wunsch II
This paper presents an adaptive resonance theory predictive mapping (ARTMAP) model which uses incremental cluster validity indices (iCVIs) to perform unsupervised learning, namely iCVI-ARTMAP.
no code implementations • 4 May 2019 • Leonardo Enzo Brito da Silva, Islam Elnabarawy, Donald C. Wunsch II
This survey samples from the ever-growing family of adaptive resonance theory (ART) neural network models used to perform the three primary machine learning modalities, namely, unsupervised, supervised and reinforcement learning.
no code implementations • 18 Feb 2019 • Leonardo Enzo Brito da Silva, Niklas M. Melton, Donald C. Wunsch II
Validation is one of the most important aspects of clustering, but most approaches have been batch methods.