Litcius/Paper detail

Mitigating severe over-parameterization in deep convolutional neural networks through forced feature abstraction and compression with an entropy-based heuristic

Nidhi Gowdra, Roopak Sinha, Stephen G. MacDonell, Wei Qi Yan

2021Pattern Recognition14 citationsDOIOpen Access PDF

Topics & Concepts

Convolutional neural networkHeuristicComputer scienceMNIST databaseHyperparameterArtificial intelligenceAbstractionFeature (linguistics)Machine learningPattern recognition (psychology)Deep learningAlgorithmLinguisticsPhilosophyEpistemologyAdvanced Neural Network ApplicationsDomain Adaptation and Few-Shot LearningHuman Pose and Action Recognition
Mitigating severe over-parameterization in deep convolutional neural networks through forced feature abstraction and compression with an entropy-based heuristic | Litcius