Prasanna Kumar Reddy Mallampati
This is a likely match — the affiliation was inferred from OpenAlex, ORCID, and web sources but has not been fully confirmed. Treat with appropriate caution.
Researcher
Unknown Researcher
Research Areas
Biography and Research Information
OverviewAI-generated summary
Prasanna Kumar Reddy Mallampati's research focuses on cybersecurity, particularly concerning advanced threat detection and defense mechanisms. His work investigates methods for identifying malicious activities within computer systems, including the analysis of volatile memory artifacts for malware classification and the detection of stealthy side-channel attacks in multi-tenant cloud environments. He also explores the vulnerabilities of deep neural networks to attacks, such as silent weight corruption affecting GPU memory, and proposes integrity defenses. Additionally, his research addresses the adversarial fragility of these networks, examining its structural causes and theoretical limits.
Mallampati collaborates with Nanda Gopal Parise at Southern Arkansas University, with whom he has co-authored four publications. His recent work, published in 2026, includes studies on hierarchical evidence-aware transformers for malware classification, GPU memory fault attacks on deep neural networks, cross-layer detection of side-channel attacks in clouds, and the adversarial fragility of deep neural networks.
Metrics
- Publications: 4
Selected Publications
-
MemFormer-H: MemFormer-H: A Hierarchical Evidence-Aware Transformer for Malware Classification Using Volatile Memory Artifacts (2026)
-
Silent Weight Corruption: Evaluating GPU Memory Fault Attacks on Deep Neural Networks and Designing Lightweight Integrity Defenses (2026)
-
Cross-Layer Detection of Stealthy Side-Channel Attacks in Multi-Tenant Clouds with Statistical False-Positive Control (2026)
-
Adversarial Fragility in Deep Neural Networks: Structural Causes, Theoretical Limits, and the Illusion of Robustness (2026)
Collaboration Network
Top Collaborators
- Adversarial Fragility in Deep Neural Networks: Structural Causes, Theoretical Limits, and the Illusion of Robustness
- Cross-Layer Detection of Stealthy Side-Channel Attacks in Multi-Tenant Clouds with Statistical False-Positive Control
- Silent Weight Corruption: Evaluating GPU Memory Fault Attacks on Deep Neural Networks and Designing Lightweight Integrity Defenses
- MemFormer-H: MemFormer-H: A Hierarchical Evidence-Aware Transformer for Malware Classification Using Volatile Memory Artifacts
Similar Researchers
Based on overlapping research topics