Dual-VAE with Truncated Gaussian: An Unsupervised Defense Against Model Poisoning in Federated Learning
Bibliographic record
Abstract
Federated learning (FL) enables decentralized model training while preserving data privacy, but remains vulnerable to model poisoning attacks, where malicious clients inject harmful updates to degrade global model performance. In this work, we propose Dual-Variational Autoencoder with truncated Gaussian (DVTG), an unsupervised defense framework for anomaly detection. The first VAE filters out low-reconstruction-error updates from potentially poisoned data. These are used to train a second VAE with a truncated Gaussian prior. This prior constrains latent representations to high-density regions of normal behavior, improving robustness against noise and adversarial manipulation. Experiments on MNIST with sign-flipping and additive Gaussian noise attacks show that our method outperforms both single-VAE and robust aggregation baselines in anomaly detection and global model performance.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.006 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".