Bibliographic record
Abstract
Federated Learning (FL) is a promising solution for training using data collected from heterogeneous sources (e.g., mobile devices) while avoiding the transmission of large amounts of raw data and preserving privacy. Current FL approaches operate in an iterative manner by selecting a subset of participants each round, asking them to training using their latest local data over the most recent version of the global model, before collecting these local model updates and aggregating them to form the next iteration of the global model, and so forth until convergence is reached. Unfortunately, existing FL approaches typically select randomly the set of clients to use each round, which can negatively impact the quality of the model trained, as well the training round time due to the straggler problem. Moreover, clients, especially mobile devices with limited resources, should be incentivized to participate as federated learning is essentially a form of crowdsourcing for AI which requires monetization. We argue that the integration of blockchain and smart contract technologies to FL can solve the two aforementioned issues. In this paper, we present FL-MAB (FL- Multi-Auction using Blockchain), a client selection mechanism for FL operating in a smart contract which rewards clients for their participation using cryptocurrencies. FL-MAB employs a multidimensional auction mechanism for selecting users based on the compute and network resources offered by each client, as well as the quality of their local data. This auction is realized in a reliable and auditable manner through a smart contract. This allows FL-MAB to measure the relative contribution of each client by calculating a Shapley value, and allocating rewards accordingly. We have implemented FL-MAB using Solidity and tested on the Ethereum blockchain with various popular datasets. Our results show that FL-MAB outperforms existing baseline schemes by improving accuracy and reducing the no. of FL rounds.
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 | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.069 | 0.233 |
| Research integrity | 0.000 | 0.001 |
| 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; both teacher heads agree on what is shown here.
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".