FATE & KubeFATE v1.8.0 released! Huge performance improvements, security enhancements, updated algorithms, optimized scheduling and testing frameworks, better high availability support and MORE!


wfangchi@...
 

Hi,

We are super excited to announce the new major release of FATE and KubeFATE - v1.8.0. After months of hard work, we have added numerous new features and improvements in this release, including performance improvements in many commonly-used components, added some most wanted features in a number of algorithms, further enhancement to the deployment and operation scenarios using KubeFATE and more. It is time to upgrade your FATE to the latest version!

Release Pages:
FATE 1.8.0: https://github.com/FederatedAI/FATE/releases/tag/v1.8.0
FATE-Flow 1.8.0: https://github.com/FederatedAI/FATE-Flow/releases/tag/v1.8.0
KubeFATE 1.8.0: https://github.com/FederatedAI/KubeFATE/releases/tag/v1.8.0

Major Features and Improvements in FATE v1.8.0

FederatedML

  • Add non-coordinated-version Hetero Linear Regression, based on integrated Hetero GLM framework, with mixed protocol of HE and SPDZ
  • Homo LR support one-vs-rest
  • Add SecureBoost-MO algorithm to speed up multi-class classification of Hetero & Homo SecureBoost, 1.5x-5x faster
  • Optimize Hetero SecureBoost Predict Transmission Data Size,reduce 75% bandwidth consumption if tree's max depth is small
  • Speed up DH Intersection implementation, 30%+ faster
  • Optimized Quantile Binning gk-summary structure & split point query,20%+ faster, less memory cost
  • Support weighted training in non-coordinated Hetero Logistic Regression & Linear Regression
  • Merge Hetero FastSecureBoost into Hetero SecureBoost as a boosting strategy option

Fate-ARCH

  • Adjustable task_cores for standalone FATE
  • Enable Eggroll option to make computing output "IN_MEMORY" by default

Fate-Test

  • Include Paillier encryption performance evaluation
  • Include SPDZ performance evaluation
  • Optimized testsuite printout
  • Include examples data upload and mnist download
  • Provide pipeline to dsl convert tools

Bug-Fix

  • Fix bug for SPDZ when using default q_filed
  • Fix multiple get problem of SPDZ
  • Fix bugs of recursive-query homo feature binning
  • Fix homo_nn's model aggregation problem
  • Fix bug for hetero feature selection when using federated filter but some party's feature is empty.

Major Features and Improvements in FATE-Flow v1.8.0

  • Optimize the model migration function to reduce user operation steps;
  • Add version compatibility check in component center to support multiple parties to use different versions;
  • Add data table disable/enable function, and support batch delete disable table

Major Features and Improvements in KubeFATE v1.8.0

KubeFATE v1.8.0 supports FATE v1.8.0 and brings the following updates:

  • Support specify ingress class names #523 ;
  • KubeFATE support liveness and readiness #552 ;
  • Adjust the file structure of the Chart to facilitate the addition of new module #526 ;
  • Solve the warning of potential security vulnerabilities in Github pages #519 ;
  • and other bug fixes #501

For the feature list details, please refer to: KubeFATE v1.8.0 Feature List


Cheers!