arXiv subjects

Suresh Kumar Amalapuram

Publications and source records attributed to Suresh Kumar Amalapuram.

4 recordsLinked to original sources

arXiv2026

SEED: Semi-supervised Continual MalwarE Detection for Tackling ConcEpt Drift on a BuDget

Machine learning based malware detectors become obsolete over time due to concept drift in benign and malware applications. Recent methods rely on fully labeled data and use hierarchical contrastive loss (HCL) with active learning to improve robustness against drift by exploiting semantic structure in malware representations. However, obtaining labeled data in the security domain is difficult. Under partially labeled settings, HCL suffers significant performance degradation in detecting unseen malware, especially on datasets such as BODMAS where strong semantic structure may not exist. In this paper, we propose SEED, a semantic-structure-agnostic method for malware detection under limited supervision. SEED combines a tailored binary cross-entropy objective with semi-supervised continual learning and active learning. For partially labeled seen tasks, unlabeled samples are projected into a representation space constructed from previously seen data using singular value decomposition, and paired with suitable labeled samples to encourage representation consistency. For unseen tasks with fully unlabeled data, uncertainty is quantified using cosine distance in representation space, and the most uncertain samples are selected for analyst labeling. We evaluate SEED on both Windows and Android malware datasets. Using only 20% labeled data on seen tasks, SEED achieves average AUT improvements of 40% on BODMAS and 14% on AndroZoo for unseen malware detection compared to HCL* (the semi-supervised adaptation of HCL), while remaining competitive on APIGraph. Finally, we introduce a delayed buffer update strategy to reduce label noise propagation during replay and improve learning stability.

cs.CR↗

arXiv2026

CITADEL: A Semi-Supervised Active Learning Framework for Malware Detection Under Continuous Distribution Drift

Android malware detection systems suffer severe performance degradation over time due to concept drift caused by evolving malicious and benign app behaviors. Although recent methods leverage active learning and hierarchical contrastive loss to address drift, they remain fully supervised, computationally expensive, and ineffective on long-term real-world benchmark. Moreover, expert labeling does not scale to the monthly emergence of nearly 300K new Android malware samples, leaving most data unlabeled and underutilized. To address these challenges, we propose CITADEL, a semi-supervised active learning framework for Android malware detection. Existing semi-supervised methods assume continuous and semantically meaningful input transformations, and fail to generalize well to high-dimensional binary malware features. We bridge this gap with malware-specific augmentations, Bernoulli bit flips and feature masking, that stochastically perturb feature to regularize learning under evolving malware distributions. \system further incorporates supervised contrastive loss to improve boundary sample discrimination and combines it with a multi-criteria active learning strategy based on prediction confidence, $L_p$-norm distance, and boundary uncertainty, enabling effective adaptation under constrained labeling budgets. Extensive evaluation on four large-scale Android malware benchmarks -- APIGraph, Chen-AZ, MaMaDroid, and LAMDA, demonstrates that \system outperforms prior work, achieving F1 score of over 1\%, 3\%, 7\%, and 14\% respectively, using only 40\% labeled samples. Furthermore, \system shows significant efficiency over prior work incurring $24\times$ faster training and $13\times$ fewer operations. \paragraph{Availability} The code is available at https://github.com/IQSeC-Lab/CITADEL.git.

cs.CR↗

arXiv2025

LAMDA: A Longitudinal Android Malware Benchmark for Concept Drift Analysis

Machine learning (ML)-based malware detection systems often fail to account for the dynamic nature of real-world training and test data distributions. In practice, these distributions evolve due to frequent changes in the Android ecosystem, adversarial development of new malware families, and the continuous emergence of both benign and malicious applications. Prior studies have shown that such concept drift -- distributional shifts in benign and malicious samples, leads to significant degradation in detection performance over time. Despite the practical importance of this issue, existing datasets are often outdated and limited in temporal scope, diversity of malware families, and sample scale, making them insufficient for the systematic evaluation of concept drift in malware detection. To address this gap, we present LAMDA, the largest and most temporally diverse Android malware benchmark to date, designed specifically for concept drift analysis. LAMDA spans 12 years (2013-2025, excluding 2015), includes over 1 million samples (approximately 37% labeled as malware), and covers 1,380 malware families and 150,000 singleton samples, reflecting the natural distribution and evolution of real-world Android applications. We empirically demonstrate LAMDA's utility by quantifying the performance degradation of standard ML models over time and analyzing feature stability across years. As the most comprehensive Android malware dataset to date, LAMDA enables in-depth research into temporal drift, generalization, explainability, and evolving detection challenges. The dataset and code are available at: https://iqsec-lab.github.io/LAMDA/.

cs.CR↗

arXiv2024

SOUL: A Semi-supervised Open-world continUal Learning method for Network Intrusion Detection

Fully supervised continual learning methods have shown improved attack traffic detection in a closed-world learning setting. However, obtaining fully annotated data is an arduous task in the security domain. Further, our research finds that after training a classifier on two days of network traffic, the performance decay of attack class detection over time (computed using the area under the time on precision-recall AUC of the attack class) drops from 0.985 to 0.506 on testing with three days of new test samples. In this work, we focus on label scarcity and open-world learning (OWL) settings to improve the attack class detection of the continual learning-based network intrusion detection (NID). We formulate OWL for NID as a semi-supervised continual learning-based method, dubbed SOUL, to achieve the classifier performance on par with fully supervised models while using limited annotated data. The proposed method is motivated by our empirical observation that using gradient projection memory (constructed using buffer memory samples) can significantly improve the detection performance of the attack (minority) class when trained using partially labeled data. Further, using the classifier's confidence in conjunction with buffer memory, SOUL generates high-confidence labels whenever it encounters OWL tasks closer to seen tasks, thus acting as a label generator. Interestingly, SOUL efficiently utilizes samples in the buffer memory for sample replay to avoid catastrophic forgetting, construct the projection memory, and assist in generating labels for unseen tasks. The proposed method is evaluated on four standard network intrusion detection datasets, and the performance results are closer to the fully supervised baselines using at most 20% labeled data while reducing the data annotation effort in the range of 11 to 45% for unseen data.

cs.CR↗