The Role of Unsupervised Learning in Defending Against Adversarial Attacks
Bibliographic record
Abstract
The fast-changing realm of machine learning and artificial intelligence makes model susceptibility to adversarial assaults a serious concern. Adversarial assaults manipulate input data to influence machine learning model estimations. This threatens self-driving vehicles, healthcare, and cybersecurity. To solve this challenge, we investigate how unsupervised learning might prevent adversarial assaults. Our paper introduces the AutoEncoder-Based Adversarial Detector (AED), Variational Autoencoders for Adversarial Feature Extraction (VAE-AFE), and Clustering and Density-Based Hybrid Defense. These strategies increase machine learning security via uncontrolled learning. Because they recreate raw data, extract strong traits, and apply clustering and density-based techniques, these methods discover and reduce adversarial hazards well. We demonstrate that these techniques outperform adversarial defensive tactics in prolonged trials. Accuracy, precision, recall, F1 Score, and ROC AUC reveal that the recommended techniques increase over time. The offered solutions offer a clear defense against dynamic threats, surpassing previous ways and securing AI applications. As artificial intelligence becomes increasingly widespread, machine learning model security and integrity are crucial. These approaches offer promise for this objective and valuable insights into adversarial defense, a burgeoning field.
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 | 0.005 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.001 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| 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".