SemiS-VFL: A Semi-Supervised Machine Learning Frameworkfor Vertical Federated Learning
Bibliographic record
Abstract
Federated Learning (FL) is a promising Machine Learning (ML) approach, that enables collaborative model training across multiple entities without centralizing or replicating raw data in a central data repository. In a Vertical Federated Learning (VFL) setting, datasets in various clients (or parties) share similar sample spaces but possess distinct features. One of the challenges of developing VFL models is the presence of samples that are represented across all parties (aligned samples), while other non-aligned samples are only represented in a subset of the clients. Recent surveys showed that solutions to address the concern of insufficient aligned training samples across all parties and imbalanced datasets in VFL are still under-explored. In this paper, we propose a novel semi-supervised machine learning approach for vertical federated learning (SemiS-VFL) that enhances the prediction performance for the party owning the labels by locally training a set of global models based on both aligned samples and all local samples. Through this designed approach, the other parties that do not have access to the label feature can creatively obtain the ability to predict it. Our extensive experimental results show that, overall, SemiS-VFL outperforms alternative solutions in accuracy and F1-score for minority classes when employing batch-balancing and pseudo-label techniques. The proposed solution presents clear advantages to dealing with VFL in an imbalanced setting.
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 | 0.009 | 0.014 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.005 |
| Open science | 0.004 | 0.005 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".