Image-Based Malware Detection Using α-Cuts and Binary Visualisation
Image conversion of malicious binaries, or binary visualisation, is a relevant approach in the
security community. Recently, it has exceeded the role of a single-file malware analysis tool …
security community. Recently, it has exceeded the role of a single-file malware analysis tool …
Quantifying lottery tickets under label noise: accuracy, calibration, and complexity
Pruning deep neural networks is a widely used strategy to alleviate the computational
burden in machine learning. Overwhelming empirical evidence suggests that pruned …
burden in machine learning. Overwhelming empirical evidence suggests that pruned …
On How Iterative Magnitude Pruning Discovers Local Receptive Fields in Fully Connected Neural Networks
Since its use in the Lottery Ticket Hypothesis, iterative magnitude pruning (IMP) has become
a popular method for extracting sparse subnetworks that can be trained to high performance …
a popular method for extracting sparse subnetworks that can be trained to high performance …
Deep Reinforcement Learning Agent for Dynamic Pruning of Convolutional Layers
Convolutional neural networks have become ubiquitous in image classification tasks. The
state-of-the-art models for image classifications use convolutional layers in one way or …
state-of-the-art models for image classifications use convolutional layers in one way or …
Understanding Vision and Language Representations under the Lens of Intrinsic Dimension
H Wang, Q Ma - openreview.net
Current multimodal representation learning is mainly achieved by intuitive and heuristic
approaches. However, the cooperation and the utility of each modality remain unclear. We …
approaches. However, the cooperation and the utility of each modality remain unclear. We …