VZVictor Zinrobustramblings.hashnode.dev·Dec 14, 2024 · 6 min readPaper summary - Impact of Aliasing on Generalization in Deep Convolutional NetworksTL; DR by carefully selecting where to put aliasing mitigation in a CNN it is possible to limit anti-aliasing information loss. Aliasing negatively impact performances in a way that is not evident from validation loss. Another paper on the effect of ...00
VZVictor Zinrobustramblings.hashnode.dev·Dec 8, 2024 · 4 min readDoes anti-aliasing smooth spatial loss surface?The last two posts (1, 2) I published were on the topic of spatial robustness, i.e. the stability of a neural network prediction with respect to small translation and rotation. A curious observation from [1] states that first order adversary seem to ...00
VZVictor Zinrobustramblings.hashnode.dev·Nov 30, 2024 · 4 min readPaper summary - Making Convolutional Networks Shift-Invariant AgainTL; DR Many operations in a typical convolutional neural network down-sample the signal without accounting for aliasing. Aliasing causes frequency content that makes a classifier vulnerable to rotation and translations. Aliasing in CNNs Aliasing is a...00
VZVictor Zinrobustramblings.hashnode.dev·Nov 24, 2024 · 4 min readPaper review - Exploring the landscape of spatial robustnessTL; DR Image classification model (probably vision models in general) are spatially brittle. Small combination of rotation and translation can radically change model output. Introduction The focus in on spatial robustness, i.e. small spatial transfor...00
VZVictor Zinrobustramblings.hashnode.dev·Nov 12, 2024 · 2 min readPaper review - A Study and Comparison of Human and Deep Learning Recognition Performance Under Visual DistortionsWhile this is an older paper (2017), it holds significant relevance in the field of robustness. The paper addresses a fundamental and still relevant question: do neural networks and human vision operate similarly? How do they handle distorted image i...00
