Adversarial attacks on anomaly detectors in process systems: A case study on Tennessee Eastman process dataset
Bibliographic record
Abstract
Abstract The integration of machine learning (ML) techniques into industrial control systems (ICS) across various industries, including chemical process plants, has revolutionized operational decision‐making by enabling faster and more informed responses. However, ICS are increasingly vulnerable to critical challenges, particularly adversarial attacks and data privacy concerns. Adversarial samples, crafted to exploit weaknesses in ML algorithms, can cause misclassifications or erroneous predictions, often evading detection by operators. Such attacks pose substantial risks to anomaly detection systems within ICS, potentially resulting in financial losses, operational disruptions, infrastructure damage, or threats to human safety. While adversarial attacks have been extensively studied in domains such as image and audio processing, their implications for industrial environments, particularly chemical processes, remain underexplored. Industrial systems heavily depend on anomaly detectors to identify deviations from normal operations, making these systems prime targets for adversarial manipulations. This work aims to address this gap by developing anomaly detection models trained on the benchmark Tennessee Eastman process (TEP) dataset. Adversarial attacks are generated and applied to these anomaly detection models to assess their impact and evaluate the performance of these models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".