Reproduce “DYNAS: A Dynamic Network Penetration Testing Framework Using Deep Hierarchical Reinforcement Learning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Md. Rakibul Hassan, Majedul Islam, Md. Ashraful Alam, Muhammad Iqbal Hossain, Swakkhar Shatabda
DYNAS extends NASimEmu with configurable defenses, noisy scans, temporary service changes, action timeouts, and staged training. Pen-DHRL uses a recurrent manager–worker policy with IDS context to select persistent subgoals. Trained without IDS and evaluated with IDS enabled, it leads reward per interaction on Bridge and Dynamic and host capture on VarA and VarB, while GNN leads other reported metrics.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.