Figure 4 - uploaded by Ghazi Gharsallah
Content may be subject to copyright.
Download
View publication
24: The feature map difference between the original data and the PGD attack data extracted from the non-robust model.

24: The feature map difference between the original data and the PGD attack data extracted from the non-robust model.

Source publication
Figure 2.3: The Sigmoid Activation Function Curve
Figure 2.4: 224 × 224 input image from [4]
Figure 2.6: Features map extracted from a pre-trained...
Figure 3.1: An application from [8] of Fast Gradient Sign Method (FGSM)...
+23 Figure 3.2: An application from [25] of Projected Gradient Descent...
AI Classifiers Robustness to Adversarial Attacks: Manipulating Output of Classifiers by Small Input Perturbations
Thesis
Full-text available
  • Jun 2020
Deep Learning has shown an advanced performance in the state of the art applications in several domains. However, its vulnerability to adversarial attacks can remarkably ruin any powerful classifier's performance through manipulating the output estimation with small perturbations, that provoked Machine Learning research community to go through this...
Cite
Download full-text

Similar publications

Comparisons of GFLOPs between RECLIP and the baseline model during the...
RECLIP with one-stage or multi-stages high-resolution finetuning. We...
RECLIP: Resource-efficient CLIP by Training with Small Images
Preprint
Full-text available
  • Apr 2023
We present RECLIP (Resource-efficient CLIP), a simple method that minimizes computational resource footprint for CLIP (Contrastive Language Image Pretraining). Inspired by the notion of coarse-to-fine in computer vision, we leverage small images to learn from large-scale language supervision efficiently, and finetune the model with high-resolution...
View
Figure 1 Product innovativeness and order of entry matrix
Figure 2 Conceptual model
Descriptive statistics and correlation matrix a
Results of post-stratification interactions (Standardised regression...
International product rollout strategies and performance: the moderating effects of product innovativeness, order of entry and firm size
Article
Full-text available
  • Jan 2023
Export marketing strategy requires firms to decide whether to launch a new product sequentially or simultaneously across targeted foreign markets. While a handful of papers have explored the strategic launch concepts, the research community has yet to empirically assess how they affect multiple performance outcomes along the strategic-fit paradigm...
View
Figure 1: Taxonomy of tasks in the literature on Knowledge Graphs in...
Figure 2: Distribution of number of papers from 2012 to 2021 (database...
Figure 3: Number of papers by most popular application domains.
Figure 4: Overview of most popular tasks by number of application domains.
+2 Figure 5: Distribution of number of papers by most popular tasks from...
A Decade of Knowledge Graphs in Natural Language Processing: A Survey
Preprint
Full-text available
  • Sep 2022
Accepted to AACL-IJCNLP 2022 / In pace with developments in the research field of artificial intelligence, knowledge graphs (KGs) have attracted a surge of interest from both academia and industry. As a representation of semantic relations between entities, KGs have proven to be particularly relevant for natural language processing (NLP), experien...
View
Fig. 1. The five categories of fuzzers that involve a grammar, an...
Fig. 2. Evolutionary fuzzers
Evolutionary grammar learner fuzzers
Fuzzers for stateful systems: Survey and Research Directions
Preprint
Full-text available
  • Jan 2023
Fuzzing is a security testing methodology effective in finding bugs. In a nutshell, a fuzzer sends multiple slightly malformed messages to the software under test, hoping for crashes or weird system behaviour. The methodology is relatively simple, although applications that keep internal states are challenging to fuzz. The research community has re...
View
Experimental condition settings.
Research on Online Monitoring Method for Bond Wire Fatigue Applied to IGBT Module
Article
Full-text available
  • Aug 2023
Bond wire fatigue is a prominent problem affecting the reliability of IGBT modules and power electronic converters, and has thus received much attention in the research community. As bond wire fatigue leads to changes in the gate circuit parameters, it affects the IGBT turn-off process. Based on this finding, this paper proposes a fatigue evaluatio...
View