---
title: "A Hidden Markov Preference Model for Online Prediction of Adversary Actions"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.08445v1"
permalink: "https://feed.astralor.com/items/itm_e436ef1ca2d820a2e79a0a3d"
published: "2026-11-05T18:37:07.000Z"
score: 58
channel: "ai"
tags: ["Hidden Markov Preference","Model","Online Prediction","Adversary Actions","Two","Markov"]
---

# A Hidden Markov Preference Model for Online Prediction of Adversary Actions

## At a glance

A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs. Two structures complicate this: the attacker's behavioral regime…

## Key points

- A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs.
- Two structures complicate this: the attacker's behavioral regime persists and can switch within an episode, and each action combines an operation with state-dependent arguments whose feasible values change over time.
- We capture both with a hidden Markov model whose latent state is the active regime and whose regime-conditioned action law is defined on a feasibility-masked tree of executable actions.

## Full text

A defender anticipating an attacker needs a probability distribution over the attacker's next action from information available before that action occurs. Two structures complicate this: the attacker's behavioral regime persists and can switch within an episode, and each action combines an operation with state-dependent arguments whose feasible values change over time. We capture both with a hidden Markov model whose latent state is the active regime and whose regime-conditioned action law is defined on a feasibility-masked tree of executable actions. The same likelihood forms the next-action distribution and updates the regime belief once the action is seen, and a partially observed prefix contributes its resolved local factors. Estimation separates into convex logistic fits that Baum-Welch reuses when regimes are unlabeled. We derive a reach-weighted nodewise KL expression for regime information, a sequence-wise log-loss bound for switching regime paths, and a uniform fallback guarantee. On the CAGE 2 cyber defense benchmark the filtered latent state is worth 0.8 nats per step, and after a controller switch the loss begins to fall in three steps against ten for a recurrent baseline.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_e436ef1ca2d820a2e79a0a3d