cs.AISep 29, 2026

Distinguish or Homogenize: Last-Chance Policy Identification and Risk-Budgeted Recovery under Irreversible Resource Depletion

Authors: Yibo Guo, Xiaodan Wang

Organizations: School of Computer and Artificial Intelligence, Zhengzhou University, Henan Zhengzhou 450001, China

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Under irreversible resource depletion, an agent can spend resources to distinguish among latent fault models, or to change the system state so that the remaining models admit a common acceptable continuation--at which point further diagnosis becomes unnecessary. This distinguish-or-homogenize principle identifies a path that existing frameworks for identification, planning, and diagnosis do not make explicit: prior formulations treat the mapping from fault models to acceptable policies as a given, whereas LCPI makes it a function of the agent's own actions. We formalize this principle through Last-Chance Policy Identification (LCPI), where correctness is evaluated at the state the agent reaches rather than at the initial state. The Last Identifiable Margin (LIM) marks the feasibility boundary between distinguishing and homogenizing. For deterministic diagnostic graphs we provide the Exact-LIM recursion; for noisy finite-horizon recovery we propose Risk-Budgeted Compatibility Planning (RBCP), which searches a compatibility-aware frontier under a hard worst-case failure constraint. Across incident recovery on abstract microservice topologies and latent-damage navigation in MiniGrid, RBCP improves risk-feasible recovery while satisfying the failure budget. A sham control--cost-matched actions that preserve model incompatibility--eliminates the gain entirely, confirming that the benefit comes from changing which policies are acceptable for which models, not from extra search or additional budget.

Figures & tables

Fig. 1 : Overview of the LCPI framework. At each step the agent chooses between distinguishing active fault models and homogenizing the decision landscape. The Last Identifiable Margin (LIM) certifies the feasibility boundary between the two pathways.
AxisEnv.ComparisonDifference95% CIpp
Det.MIRRBCP vs. RC-POMDP+0.0067[0.0028,0.0101]0.0010
Det.MLDRBCP vs. Minimax+0.3512[0.3359,0.3673]0.0005
ShockMIRRBCP vs. Minimax+0.1553[0.1499,0.1607]0.0005
ShockMLDRBCP vs. Minimax+0.1558[0.1351,0.1776]0.0005
TABLE I : Locked-test paired comparisons in WM-S-AUC. Positive values favor RBCP. All comparisons pass the noninferiority gate.

Explore similar work

CardsList
Oct 1, 2026cs.LG

Exact Distinguishability in Non-Markovian Decision Processes

Non-Markovian environments are often modeled as Regular Decision Processes (RDPs), where dynamics depend on the interaction history through a finite automaton. Existing offline guarantees for RDPs rely on a distinguishability assumption on the behaviour policy but provide no means of verifying it. When the assumption is violated, distinct models may explain the data equally well. We study when data collected under a fixed behaviour policy can distinguish two candidate RDPs. We prove that the posterior odds between observationally equivalent candidates remain equal to the prior odds at every sample size, even when the policy visits every automaton state, and verify both results formally in Lean 4. We then characterize this equivalence exactly and derive PEC, an algorithm that decides it in time linear in the size of the product automaton. The distinguishability assumption of prior work fails on three of our four test environments, and the experiment identified by PEC restores it in each case.

Kabir Murjani, Nisarg Patel

Nirma University Ahmedabad, India · Google Mountain View, CA

Markov Decision ProcessesIndistinguishabilityAutomata
AskPDF
May 11, 2026cs.DC

An Uncertainty-Aware Resilience Micro-Agent for Causal Observability in the Computing Continuum

Grey failures in the computing continuum produce ambiguous overlapping symptoms that existing approaches fail to diagnose reliably, either due to a lack of causal awareness or acting under high epistemic uncertainty, risking destructive interventions. This paper presents an uncertainty-aware resilience micro-agent for causal observability (AURORA), a lightweight framework for diagnosing and mitigating grey failures in edge-tier environments. The framework employs parallel micro-agents that integrate the free-energy principle, causal do-calculus, and localized causal state-graphs to support counterfactual root-cause analysis within each fault's Markov blanket. Restricting inference to causally relevant variables reduces computational overhead while preserving diagnostic fidelity. AURORA further introduces a dual-gated execution mechanism that authorizes remediation only when causal confidence is high and predicted epistemic uncertainty is bounded; otherwise, it abstains from local intervention and escalates the diagnostic payload to the fog tier. Our experiments demonstrate that AURORA outperforms baselines, achieving a 0% destructive action rate, while maintaining 62.0% repair accuracy and a 3ms mean time to repair.

Suvi De Silva, Alfreds Lapkovskis, Alaa Saleh +2

Department of Computer Systems and Sciences, Stockholm University, Stockholm 164 25, Sweden · Department of Computer Science, University of Helsinki, Helsinki 00014, Finland

Partial ObservabilityContinuum
AskPDF
Jul 22, 2026cs.AI

Safe Remediation as Risk-Constrained Intervention Decision in Microservice Systems

In modern IT operations (IT-Ops), the cost of an incorrect repair often exceeds the cost of no action at all. Yet existing automated remediation systems are designed to generate actions rather than to decide whether intervention is warranted, leaving safety as an afterthought enforced by manual approval. This paper makes three contributions to close this gap: (i) we reformulate safe remediation as a risk-constrained intervention decision problem and cast it as a Constrained Markov Decision Process (CMDP), in which the agent maximizes repair success subject to a bounded false remediation rate (FRR); (ii) we introduce a three-dimensional risk decomposition comprising blast radius, reversibility, and epistemic uncertainty, providing operators with an interpretable per-action safety interface; and (iii) we design a context-adaptive human-in-the-loop (HITL) gate that turns escalation from a binary failsafe into a bandwidth-aware control layer responsive to on-call load and business criticality. The full policy is learned offline from historical incident logs, enabling explicit control of the expected FRR. Experiments on the Train Ticket microservice benchmark with Chaos Mesh fault injection and an RCAEval-aligned fault taxonomy show that our framework reduces FRR by 39% while improving repair success by 2.5 points over a strong runbook baseline, and reduces on-call escalation load by 17% relative to a fixed-threshold variant.

Chengxiao Dai, Zhaokun Yan, Chenjun Lei +2

School of Computer Science, University of Sydney, Sydney, Australia · School of Computation, Information and Technology, Technical University of Munich, Munich, Germany · College of Business and Economics, Australian National University, Canberra, Australia +2

RemediationMicroservicesConformal Risk ControlReal-World Cloud Fault Injection Dataset
AskPDF