Sign in Subscribe
Cloud computing

Cloud Computing Security Threats: New Research on Deep Reinforcement Learning for DDoS Detection

Strategic Insights

— 1 min read

Researchers from Siksha 'O' Anusandhan University have proposed a deep reinforcement learning (DRL)-based framework for real-time Distributed Denial of Service (DDoS) detection in cloud environments. The study, published in Scientific Reports, investigates the effectiveness of three actor-critic DRL algorithms: Twin Delayed Deep Deterministic Policy Gradient (TD3), Deep Deterministic Policy Gradient (DDPG), and Advantage Actor-Critic (A2C) in differentiating between benign and malicious network traffic. The proposed approach combines Boruta, SHAP, and cross-validation stability analysis for robust feature selection and achieves superior performance in experimental results.

Key Takeaways:

  • The proposed DRL-based approach overcomes the limitations of traditional Intrusion Detection Systems (IDSs) and provides an adaptive solution for DDoS detection in dynamic cloud environments.
  • The study investigates the effectiveness of three actor-critic DRL algorithms: TD3, DDPG, and A2C in real-time DDoS detection.
  • The proposed approach combines Boruta, SHAP, and cross-validation stability analysis for robust feature selection and achieves superior performance in experimental results.
  • The TD3 algorithm demonstrates superior performance, with an average accuracy of 99.12%, an AUC of 99.21%, and an inference latency of 1.87 milliseconds per sample.
  • An ablation study confirms the critical contribution of each preprocessing component, and SHAP-based analysis is employed to interpret model decisions by identifying key traffic features influencing predictions.
  • The findings underscore the effectiveness, scalability, and interpretability of the proposed DRL-based approach in overcoming the limitations of traditional IDSs and providing an adaptive solution for DDoS detection in dynamic cloud environments.
  • The research team consists of Suneeta Satpathy, Centre for Cybersecurity, Siksha 'O' Anusandhan University, Bhubaneswar, Odisha, India, and additional authors Uttpal Tripathy and Pratik Kumar Swain.

Statistics:

  • Average accuracy of TD3 algorithm: 99.12%
  • AUC of TD3 algorithm: 99.21%
  • Inference latency of TD3 algorithm: 1.87 milliseconds per sample
  • Ablation study confirms the critical contribution of each preprocessing component.
  • SHAP-based analysis identifies key traffic features influencing predictions.

Sources:

  • ScienceDirect: Cloud-based DDoS detection using hybrid feature selection with deep reinforcement learning (DRL). Scientific Reports, 2025;15(1):36546. (Contact Nature Portfolio, Heidelberger Platz 3, Berlin, 14197, Germany)
  • Nature Portfolio: www.nature.com/
  • Scientific Reports: www.nature.com/srep/
  • NewsRx LLC: NewsRx. Research Conducted at Siksha 'O' Anusandhan University Has Provided New Information about Cloud Computing [Cloud-based DDoS detection using hybrid feature selection with deep reinforcement learning (DRL)]. Information Technology Newsweekly. November 4, 2025; p 667.

Read more