129 citations · 718 across the 31 of their papers we have counts for
7 papers · 1 filter
Rotated Ring, Radial and Depth Wise Separable Radial Convolutions
Wolfgang Fuhl, Enkelejda Kasneci
Simple image rotations significantly reduce the accuracy of deep neural networks. Moreover, training with all possible rotations increases the data set, which also increases the tr…
Weight and Gradient Centralization in Deep Neural Networks
Wolfgang Fuhl, Enkelejda Kasneci
Batch normalization is currently the most widely used variant of internal normalization for deep neural networks. Additional work has shown that the normalization of weights and ad…
Explainable Online Validation of Machine Learning Models for Practical Applications
Wolfgang Fuhl, Yao Rong, Thomas Motz +4
We present a reformulation of the regression and classification, which aims to validate the result of a machine learning algorithm. Our reformulation simplifies the original proble…
Multi Layer Neural Networks as Replacement for Pooling Operations
Wolfgang Fuhl, Enkelejda Kasneci
Pooling operations, which can be calculated at low cost and serve as a linear or nonlinear transfer function for data reduction, are found in almost every modern neural network. Co…
Fully Convolutional Neural Networks for Raw Eye Tracking Data Segmentation, Generation, and Reconstruction
Wolfgang Fuhl, Yao Rong, Enkelejda Kasneci
In this paper, we use fully convolutional neural networks for the semantic segmentation of eye tracking data. We also use these networks for reconstruction, and in conjunction with…
Differential Privacy for Eye Tracking with Temporal Correlations
Efe Bozkir, Onur Günlü, Wolfgang Fuhl +2
New generation head-mounted displays, such as VR and AR glasses, are coming into the market with already integrated eye tracking and are expected to enable novel ways of human-comp…
Reinforcement learning for the privacy preservation and manipulation of eye tracking data
Wolfgang Fuhl, Efe Bozkir, Enkelejda Kasneci
In this paper, we present an approach based on reinforcement learning for eye tracking data manipulation. It is based on two opposing agents, where one tries to classify the data c…