Welcome to IBM Federated Learning

IBM Federated Learning is a framework for federated learning in an enterprise environment. Federated Learning (FL) is a distributed machine learning process in which each participant node (or party) retains in data locally and interacts with the other participants via a learning protocol. The main driver behind FL is need to not share data with others, mainly driven by privacy and confidentially concerns.

IBM Federated Learning provides a basic fabric for FL on which advanced features can be added. It is not dependent on any specific machine learning framework and supports different learning topologies, e.g., an shared aggregator, and protocols. This is a team effort of multiple IBM Research teams in Almaden, Yorktown Heights, and Dublin. It is meant to provide a solid basis for federated learning that enables a large variety of federated learning models, topologies, learning models etc., in particular in enterprise settings.

Federated Learning is still in its infancy. Many aspects of machine learning have to be re-thought when not sharing data, including hyper-parameter tuning, privacy of models, optimization sensitive to privacy budgets and many more. IBM Federated Learning provides a robust foundation for this advanced work, and a target to which you can contribute your threshold homomorphic encryption aggregator using secure multi-party encryption protocol or your noise and clipping functions for the implementation of differential privacy.

IBM Federated Learning supports the following out-of-the-box model implementations at present:

• Neural networks (any neural network topology supported by Keras)

• Linear classifiers/regressions (with regularizer): logistic regression, linear SVM, ridge regression and more

• Decision Tree ID3

• Deep Reinforcement Learning algorithms including DQN, DDPG, PPO and more

• Naïve Bayes

Multiple fusion algorithms and advanced privacy features are also supported.

Modules

• ibmfl.aggregator
  • Coordinate-Median Aggregation Fusion
  • Decision Tree Fusion
  • Federated Averaging Fusion
  • Base Class
  • Gradient-Based Fusion
  • Iterative Averaging Fusion
  • Krum Fusion
  • Gaussian Naive Bayes Fusion
  • PFNM Aggregation Fusion
  • RL Averaging Fusion
  • Weight-Based RL Averaging Fusion
  • SPAHM Aggregation Fusion
  • Zeno Fusion
  • Protohandler Base Class
  • Aggregator
  • FL Metrics
  • Party Connection
  • Aggregator States
• ibmfl.connection
  • FL Connection
  • Flask Connection
  • Router
• ibmfl.data
  • Base Class
  • Reinforcement Learning Environment Handler
  • Environment Handler Base Class
  • Pandas Data Handler
• ibmfl.message
  • JSON Serializer
  • Message
  • Message Type
  • Pickle Serializatiom
  • Abstract Serializer
  • Serialization Factory
  • Serializer Types
• ibmfl.model
  • Base Class
  • Custom RL Wrapper
  • Decision Tree
  • Keras Wrapper
  • Model Update Dictionary
  • Differentially Private Gaussian Naive Bayes Wrapper
  • RL Import Wrapper
  • Sklearn Wrapper
  • Sklearn K Means Wrapper
  • SGD Classifier Wrapper
• ibmfl.party
  • Base Class
  • Federated Averaging Local Training
  • Gradient-Based Local Training
  • PFNM Local Training
  • RL Local Training
  • Party
  • Party Protocol Handler
  • Status Type
• ibmfl.storage
  • Data Storage
  • File Storage
• ibmfl.util
  • Abstract Agent
  • Abstract Replay Memory
  • Abstract Rollout
  • DQN Agent
  • DDPG Agent
  • Critic Network
  • Actor Network
  • Ornstein-Ulhenbeck Process
  • Pendulum DDPG Rollout
  • Pendulum Replay Experience
  • Adult Decision Tree Data
  • Adult Sklearn Data
  • Airline Data
  • Binovf Data
  • Diabetes Data
  • Federated Clustering Data
  • Higgs Data
  • Linovf Data
  • MNIST Keras Data
  • MNIST Sklearn Data
  • Multovf Data
  • Nursery Data
  • Pendulum Environment
  • Pendulum Environment Data
  • Rollout Save

Indices and tables

• Index

• Module Index

• Search Page