A Review of Lightweight Multi-Party Computation and Federated Learning in Financial Systems
Bibliographic record
Abstract
As the financial services are increasingly moving to the edge devices, safeguarding these sensitive transaction data without compromising on privacy has become a challenge. This systematic literature review analyzes peer-reviewed studies that explore lightweight federated learning (FL) techniques, gradient compression methods, and secure multi-party computation (MPC) protocols for privacy-preserving machine learning in financial systems. Our findings show that approaches like Federated Dropout (FedDrop), Quantized Stochastic Gradient Descent (QSGD), and Sparse Ternary Compression (STC) have been proposed to address communication overhead and device constraints. Furthermore, the review highlights privacypreserving frameworks having secure aggregation along with exploring lightweight MPC for distributed financial devices. Even with these advances, the real-world deployments within financial environments remain limited and act as gaps in privacy, communication cost, and model accuracy, especially for non-IID data. This timely review outlines the future research directions, like adaptive model compression and scalable secure aggregation frameworks for heterogeneous financial systems.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 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".