Research map / Continual learning / Online and replay-based
Augmented Memory Replay-based Continual Learning Approaches for Network Intrusion Detection
Suresh kumar Amalapuram, Sumohana S. Channappayya, Bheemarjuna Tamma
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use ECBRS (global class statistics) instead of CBRS for memory population | replay memory · SVHN, CIFAR-10, CIFAR-100, CLEAR-10/100, AnoShift | PR-AUC (A): 7% (avg) | Improved |
| use ECBRS instead of CBRS | replay memory · CIFAR-10 | PR-AUC (A): 0.883 -> 0.953 | Improved |
| use ECBRS instead of CBRS | replay memory · CIFAR-100 | PR-AUC (A): 0.572 -> 0.663 | Improved |
| use ECBRS instead of CBRS | replay memory · CLEAR-10 | PR-AUC (A): 0.941 -> 0.937 | No clear effect |
| use ECBRS instead of CBRS | replay memory · AnoShift | PR-AUC (A): 0.949 -> 0.949 | No clear effect |
| add ECBRS as memory population to MIR | MIR · all benchmarks | PR-AUC (A): 12% (avg) | Improved |
| add ECBRS to MIR on large-scale benchmarks | MIR · AnoShift, CICIDS-2018, UNSW-NB15, CICIDS-2017 | PR-AUC (A): 30% (avg) | Improved |
| add ECBRS to MIR | MIR · AnoShift | PR-AUC (A): 0.655 -> 0.942 | Improved |
| replace virtual SGD updates in MIR with PAPA (GMM perturbation) | MIR · CICIDS-2017 | training time (s): 316.0 -> 188.8 | Improved |
| replace virtual SGD updates in MIR with PAPA | MIR · CICIDS-2018 | training time (s): 9040.0 -> 5948.0 | Improved |
| replace virtual SGD updates in MIR with PAPA | MIR · KDDCUP'99 | PR-AUC (B): 0.929 -> 0.920 | Hurt |
| replace virtual SGD updates in MIR with PAPA | MIR · CLEAR-10 | PR-AUC (A): 0.953 -> 0.943 | Hurt |
Closest papers by meaning
- MalCL: Leveraging GAN-Based Generative Replay to Combat Catastrophic Forgetting in Malware Classification (AAAI 2025 · similarity 0.84)
- One-for-More: Continual Diffusion Model for Anomaly Detection (CVPR 2025 · similarity 0.83)
- Unsupervised Continual Anomaly Detection with Contrastively-Learned Prompt (AAAI 2024 · similarity 0.83)
- Attack on Prompt: Backdoor Attack in Prompt-Based Continual Learning (AAAI 2025 · similarity 0.83)
- Revisiting Generative Replay for Class Incremental Object Detection (CVPR 2025 · similarity 0.82)
- Online Adaptive Anomaly Thresholding with Confidence Sequences (ICML 2024 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
