Comparative Analysis of Imbalanced Malware Byteplot Image Classification using Transfer Learning

TL;DR

The work addresses the existence and nontriviality of $T$-periodic orbits for convex Hamiltonian systems with subquadratic growth at infinity by formulating the problem as a boundary-value problem $\dot{x}=JH'(t,x)$, $x(0)=x(T)$, and applying a variational dual-action framework. Central to the results are spectral-gap criteria involving $\gamma={\rm smallest\ eigenvalue\ of}\ B_{\infty}-A_{\infty}$ and $\lambda={\rm largest\ negative\ eigenvalue\ of}\ J\frac{d}{dt}+A_{\infty}$, with existence guaranteed when $\lambda+\gamma<0$, and nontriviality when $\gamma<-\lambda<\delta$ for a finite $\delta$. The analysis yields not only existence but also nonconstant periodic solutions and subharmonics, under both autonomous and nonautonomous (through forcing terms) settings, extending classical results by Rabinowitz, Clarke–Ekeland, and subsequent work by Michalek and Tarantello. These results provide a rigorous variational route to understand periodic dynamics in convex Hamiltonian systems with subquadratic growth, with implications for multiplicity and minimal period of orbits.

Abstract

Cybersecurity is a major concern due to the increasing reliance on technology and interconnected systems. Malware detectors help mitigate cyber-attacks by comparing malware signatures. Machine learning can improve these detectors by automating feature extraction, identifying patterns, and enhancing dynamic analysis. In this paper, the performance of six multiclass classification models is compared on the Malimg dataset, Blended dataset, and Malevis dataset to gain insights into the effect of class imbalance on model performance and convergence. It is observed that the more the class imbalance less the number of epochs required for convergence and a high variance across the performance of different models. Moreover, it is also observed that for malware detectors ResNet50, EfficientNetB0, and DenseNet169 can handle imbalanced and balanced data well. A maximum precision of 97% is obtained for the imbalanced dataset, a maximum precision of 95% is obtained on the intermediate imbalance dataset, and a maximum precision of 95% is obtained for the perfectly balanced dataset.

Figures (2)

• Figure 1: This is the caption of the figure displaying a white eagle and a white horse on a snow field
• Figure 2: This is the caption of the figure displaying a white eagle and a white horse on a snow field

Theorems & Definitions (16)

• proposition 1
• proof
• corollary 1
• proof
• lemma 1
• proof
• theorem 1: Ghoussoub-Preiss
• definition 1
• proposition 2
• proof