7 citations · 7 across the 2 of their papers we have counts for
5 papers
On Data Augmentation and Adversarial Risk: An Empirical Analysis
Hamid Eghbal-zadeh, Khaled Koutini, Paul Primus +5
Data augmentation techniques have become standard practice in deep learning, as it has been shown to greatly improve the generalisation abilities of models. These techniques rely o…
ReLU Code Space: A Basis for Rating Network Quality Besides Accuracy
Natalia Shepeleva, Werner Zellinger, Michal Lewandowski +1
We propose a new metric space of ReLU activation codes equipped with a truncated Hamming distance which establishes an isometry between its elements and polyhedral bodies in the in…
A Novel Fibonacci Pattern in Pascal's Triangle
Bernhard Moser
The Fibonacci sequence is obtained as weighted sum along the rows in the Pascal triangle by choosing a periodic up-and-down pattern of weights from the set $\{-1,-\frac{1}{2},0, \f…
Estimating the Signal Reconstruction Error from Threshold-Based Sampling Without Knowing the Original Signal
Bernhard Moser
The problem of estimating the accuracy of signal reconstruction from threshold-based sampling, by only taking the sampling output into account, is addressed. The approach is based…
On Quasi-Isometry of Threshold-Based Sampling
Bernhard Moser
The problem of isometry for threshold-based sampling such as integrate-and-fire (IF) or send-on-delta (SOD) is addressed. While for uniform sampling the Parseval theorem provides i…