A Blockchain-based Decentralized Federated Learning Framework wth Committee Consensus

A Blockchain-based Decentralized Federated Learning Framework with Committee Consensus

基于区块链的具有委员会共识的分散联邦学习框架

模型优点

1. High efficiency: only a few nodes will validate the updates, rather than broadcasting to every node and reach an agreement.
2. K-fold：cross-validation: the committee members will not participate in the local training in the round. Therefore, the local data of the committee are taken as a validation set. As the alternating of committee members at each round, the validation set changes as well.In this setting, k-fold cross-validation on FL achieved.
3. Anti-malevolence: based on the validation scores, the corresponding nodes with better performance will be elected by the smart contract and constitute the new committee for the next training round. which means the selected local data distribution is gregarious and the node is not malicious.

1)高效: 只有少数节点会验证更新，而不是广播到每个节点并达成一致。
2) K倍交叉验证: 委员轮次不参加本地培训。因此，委员会的本地数据被视为验证集。随着每一轮委员会成员的交替，验证集也会发生变化。在这种情况下，在FL上实现了k倍交叉验证。
3)反恶意: 基于验证分数，智能合约将选出相应的表现较好的节点，组成下一轮培训的新委员会。这意味着所选择的本地数据分布是群居的，并且该节点不是恶意的。

第二部分
II. RELATED WORK
简要说明其他论文提出的方法以及缺点。

第三部分
III. THE PROPOSED FRAMEWORK
本文提出的框架

在每一轮训练中，服务器将当前模型广播给一些参与节点。

一个随机初始化的模型被放入#0块，然后第0轮训练开始。

Nodes access the current model and execute local training, and put the
verified local gradients to new update blocks.

节点访问当前模型并执行局部训练，并将验证后的局部梯度放入新的更新块。
当持续有足够多的更新块时，智能合约触发聚合，并生成下一轮的新模型并将其放在链上

We should note that the FL training only relies on the latest model block, and the historical block is stored for failure fallback and block verification

每一次更新都添加到链上是用于故障回退和验证

模型块包括：头、轮数t，全局模型
更新块包括： 头，轮数t，本地更新梯度，上传者地址，更新分数

a few honest nodes will constitute a committee in charge of verification of local gradients and blocks generation. In the meantime, the rest nodes execute local training and send the local updates to the committee. The committee then validates the updates and assign a score on them. Only the qualified updates will be packed onto the blockchain

选出几个诚实结点构成委员会，其他节点在本地训练模型，将本地模型更新发给委员会，委员会验证打分，合格的更新打包记录到区块链上。
在下一轮开始时，会根据上一轮的节点分数选出新的委员会，该委员会不会再次当选

the committee members validate the local updates by treating their data as a validation set, and the validation accuracy becomes the score. This is the minimized approach that acquires no further operation of the committee, but only the basic ability to run the learning model. After combining the scores from the various committee members, the median will become the score of this update.

委员会成员的本地数据做为测试数据，验证其他节点的本地更新并且打分。综合各委员会成员的得分后，中位数将成为本次更新的得分。

设备加入联邦训练前先被验证评估，该验证处于黑名单模式:如果设备因不当行为(如提交误导性更新、传播私人模型)被踢出社区，设备将被拒绝。

每台加入训练的设备都可以快速获取最新模型，即使未提交更新，为了激励参与者，设定如下奖励机制：
许可费:每台设备都要为全局模型的访问许可付费，这些费用由管理者保管。节点可以无限制地访问社区中的最新模型。
利润分享:每轮汇总后，管理人员根据提交更新的分数将奖励分配给相应的节点