• M. Amareshwar Sai
• Kewal Ajaykumar Dharamshi
• K. Krithik Bose
• Vishal Rathod
• Padmavathi S

Modern enterprises today have networks that are vulnerable to quickly changing threats, face attacks where thousands of discovered vulnerabilities interact with each other in a system, and thus need both probabilistic risk assessments, combined with adaptive planning for possible courses of action.Traditional vulnerability scoring models (such as CVSS and EPSS) provide potentially useful benchmarks, but fail to capture the evolution or propagation of attack scenarios across interconnected network assets.In this paper, we examine early work on network-level vulnerability assessment frameworks involving everything from deterministic scoring and adaptive learning to stochastic modeling, and we outline a broad pattern of failure and identify research gaps that tie together the two essential activities of risk measurement and decision making.To help resolve these gaps, we present a Hybrid Absorbing Markov Chain–Markov Decision Process (AMC–MDP) framework that estimates long-term compromise probabilities while learning optimal defensive actions over time.The model displays certain structural features based on graph paths and incorporates risk, cost, and reward into a unified action decision loop.The work synthesizes earlier literature that indicates this sort of hybridization can bridge the analytical transparencies needed for risk measures and adaptive controls, which can provide the basis for scaling up economically sustainable, interpretable cyber defense approaches.

Vulnerability Assessment, Attack Graphs, Markov Decision Process (MDP), Absorbing Markov Chain (AMC), Reinforcement Learning, Cybersecurity Automation

"A Survey on AI-Enhanced Network Vulnerability Assessment and the Hybrid AMC-MDP Framework", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 11, page no.e253-e260, November-2025, Available :http://www.jetir.org/papers/JETIR2511432.pdf

"A Survey on AI-Enhanced Network Vulnerability Assessment and the Hybrid AMC-MDP Framework", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 11, page no. ppe253-e260, November-2025, Available at : http://www.jetir.org/papers/JETIR2511432.pdf