Candidate-Fate Accounting for Transparent Sensor Diagnostic Pipeline Search

Haotao Xie; Yutian Chen; Yangqi Liu; Xiaoyu Jiang

cs.AIAug 19, 2026

Candidate-Fate Accounting for Transparent Sensor Diagnostic Pipeline Search

Authors: Haotao Xie, Yutian Chen, Yangqi Liu, Xiaoyu Jiang

Organizations: Hangzhou International Innovation Institute, Beihang University Hangzhou, China

Abstract

Industrial sensor diagnostics relies on preprocessing, representation, and classification pipelines, making automated pipeline search useful for reducing manual design cost. However, existing automated machine/deep learning (AutoML/AutoDL) reports typically retain only fitted trials, scores, and winners, omitting generated candidates that are invalid, pruned, skipped, cached, or unfitted. This omission limits reviewers' ability to check signal constraints, budget use, and unevaluated legal alternatives. To address this, we propose candidate-fate accounting, a candidate-level audit framework for diagnostic search traces. It records each observed candidate as auditable evidence: hashes merge repeated observations, legality checks flag invalid candidates, allocation rationales explain budget decisions, and a closed fate ledger assigns one terminal fate to each candidate. Experiments on three bearing-diagnostic datasets show that the framework detects invalid candidates and identifies 30--41 candidates omitted by fitted-trial-only reports, with closed fate records verifying complete candidate accounting while maintaining competitive diagnostic performance. The code is available at https://github.com/XXIE999/candidate-fate-accounting.

Figures & tables

Figure 1. Motivating comparison among expert-designed workflows, AutoML/AutoDL search, and candidate-fate accounting. A three-part diagram contrasts expert modeling, AutoML/AutoDL search, and candidate-fate accounting. Expert modeling is shown as interpretable and checkable but manually costly; AutoML/AutoDL is shown as automated but limited in search-process visibility; candidate-fate accounting is shown as tracking generated candidates through legality checks, allocation rationales, and final fates.

Figure 2. Overview of candidate-fate accounting. Typed legality checks classify generated candidates; a replaceable allocator prioritizes eligible ones; and a hash-based ledger consolidates repeats and assigns one terminal fate—including non-fitted outcomes—to each observed candidate. The closure condition $\Delta_{\mathrm{close}}=0$ verifies complete, non-overlapping accounting. A three-stage candidate-fate accounting workflow. First, a generated diagnostic program is decomposed into preprocessing, signal-representation, and estimator components and checked for input/output compatibility and semantic validity. Second, a budget allocator uses replaceable search strategies to prioritize generated, valid, uncached candidates that have not received a terminal fate. Third, normalized candidate hashes merge repeated observations into a ledger that assigns one of six fates: invalid, duplicate or cache, pruned, skipped, full-evaluated, or residual legal-unfitted. A zero closure gap yields auditable search evidence.

Dataset	Unique Generated	Invalid	Pruned	Skipped	Duplicate/ Cache	Full Evaluated	Residual Legal- Unfitted	Closure Gap
CWRU	$49.7{\pm}0.5$	$0{\pm}0$	$6.7{\pm}0.9$	$23.0{\pm}2.8$	$0{\pm}0$	$20.0{\pm}2.9$	$0{\pm}0$	0
Ottawa	$49.7{\pm}0.5$	$0{\pm}0$	$20.3{\pm}2.9$	$18.7{\pm}4.7$	$2.3{\pm}0.9$	$8.3{\pm}2.1$	$0{\pm}0$	0
SEU	$49.7{\pm}0.5$	$0{\pm}0$	$6.3{\pm}2.1$	$23.7{\pm}3.1$	$4.3{\pm}2.1$	$15.3{\pm}2.5$	$0{\pm}0$	0

Table 1. Closed

B=50

GLS ledger counts under typed legal search, averaged over three seeds.

Dataset	Method	Best F1 $\leq$ 30 $\uparrow$	Target Success	Evaluations to Target $\downarrow$
CWRU	RLS	0.9942 $\pm$ 0.0009	3/3	26.00 $\pm$ 7.87
CWRU	TPE	0.9982 $\pm$ 0.0009	3/3	19.00 $\pm$ 4.32
CWRU	GLS	0.9982 $\pm$ 0.0002	3/3	7.00 $\pm$ 4.55
Ottawa	RLS	0.9673 $\pm$ 0.0063	3/3	9.00 $\pm$ 5.89
Ottawa	TPE	0.9639 $\pm$ 0.0038	1/3	3.00 $\pm$ 0.00
Ottawa	GLS	0.9856 $\pm$ 0.0026	3/3	3.67 $\pm$ 1.70

Table 2. Early allocation under a shared typed legal space. Bold marks the best F1 and, among 3/3-success rows, the fewest Evaluations to Target.

Figure 3. Best validation macro-F1 within 30 full evaluations versus Evaluations to Target; upper-left is better. Three panels for CWRU, Ottawa, and SEU plot the best validation macro-F1 within 30 full evaluations against Evaluations to Target. Gray circles denote random legal search, orange squares denote Tree-structured Parzen Estimator-style sampling, and blue triangles denote guided ledger search. Guided ledger search lies in the high-F1, low-evaluation region in all three panels.

Method	CWRU F1 $\uparrow$	Ottawa F1 $\uparrow$	SEU F1 $\uparrow$
RLS	0.9951(0.0004)	0.9851(0.0046)	0.7130(0.0514)
TPE	0.9978(0.0028)	0.9691(0.0119)	0.7128(0.0242)
GLS	0.9968(0.0000)	0.9862(0.0032)	0.7341(0.0134)
Statistical + RF	0.8200	0.6216	0.0748
Spectral classical	0.7068	0.6181	0.0593
Envelope classical	0.5364	0.5926	0.5899

Table 3. Protocol-scoped final-test macro-F1; search rows show three-seed mean (standard deviation), with the best matched search policy per dataset in bold.

Algorithmic Auditing

Algorithmic auditing evaluates whether automated systems are fair, reliable, robust, compliant, and accountable in their intended settings.

See all 143 papers

Agentic4PDM: Host-Governed Evidence Admissibility for Industrial Fault Diagnosis
Oct 13, 2026Nam V. Nguyen, Hoang Van Truong, Hieu D. D. +3Industrial Fault Diagnosis Agentic Reasoning
FPT Corporation · Independent Researcher · Aalborg University
A Multi-level Information Integration Framework for Physically Verifiable Fault Diagnosis of Rotating Machinery
Jul 24, 2026Yuntong Chen, Jianyu Liu, Yingqi Li +5Predictive Maintenance Fault Detection
School of Mechanical Engineering, Northwestern Polytechnical University, Xi’an, 710072, China · Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 117543, Singapore · School of Mechanical Engineering and Automation, Beihang University, Beijing, 100191, China
Label-free Industrial Fault Detection via Adversarial Inverse Reinforcement Learning: A System for Run-to-Failure Prognostics
Jul 25, 2026Dhiraj Neupane, Mohamed Reda Bouadjenek, Richard Dazeley +1Reinforcement Learning Industrial Anomaly Detection
Deakin University · Waurn Ponds, Victoria, Australia

Oct 13, 2026J/K move · Enter open · S save

Agentic4PDM: Host-Governed Evidence Admissibility for Industrial Fault Diagnosis

Nam V. Nguyen, Hoang Van Truong, Hieu D. D., Duy Khuong Nguyen+2

FPT Corporation · Independent Researcher · Aalborg University

Industrial Fault Diagnosis Agentic Reasoning AI Agent Governance LLM Agent Reliability Bearing Fault Diagnosis

Open paper AskPDF arXiv

We study industrial fault diagnosis as a classification task that assigns sensor time-series windows to predefined health states or fault types. Scarce labels and changing operating conditions make repeated training of specialized classifiers costly. Yet structures such as periodicity, trends, and transients can remain diagnostically informative and be examined through reusable analytical tools. LLM agents can select and interpret these tools, but fixed rules or training-derived guidance may not resolve each signal's ambiguity, while unconstrained interpretation may overstate weak or conflicting results. We introduce Agentic4PDM, a framework that separates adaptive investigation from control over evidence use. LLM agents compare each signal with labeled references from every candidate class, identify unresolved distinctions, and propose follow-up measurements. A deterministic software controller, called the host, enforces contract-checked gated adjudication (CCGA): it checks requests, requires commitment to a complete tool batch before results are observed, and assigns evidence roles based on tool checks against labeled references. These roles distinguish corroborative support from advisory information. The Judge Agent proposes a diagnosis, while the controller checks protocol compliance and records the investigation. Agentic4PDM supports training-free diagnosis using pretrained representations and labeled references without task-specific diagnostic models, and a standard setting that additionally investigates trained-model predictions. Experiments on selected bearing and motor diagnosis challenges and an unfiltered industrial test set supplied by our customer show gains in aggregate macro F1 over the evaluated diagnostic and agentic baselines. Ablations further support the value of the governed workflow.

Abstract

Algorithmic Auditing

Explore similar work

Agentic4PDM: Host-Governed Evidence Admissibility for Industrial Fault Diagnosis

A Multi-level Information Integration Framework for Physically Verifiable Fault Diagnosis of Rotating Machinery

Label-free Industrial Fault Detection via Adversarial Inverse Reinforcement Learning: A System for Run-to-Failure Prognostics