IBM Federated Learning: an Enterprise Framework White Paper V0.1 (2020-07-22T00:00:00.000000Z)

TL;DR

IBM Federated Learning enables data scientists to expand their scope from centralized to federated machine learning, minimizing the learning curve at the outset while also providing the flexibility to deploy to different compute environments and design custom fusion algorithms.

Abstract

Federated Learning (FL) is an approach to conduct machine learning without centralizing training data in a single place, for reasons of privacy, confidentiality or data volume. However, solving federated machine learning problems raises issues above and beyond those of centralized machine learning. These issues include setting up communication infrastructure between parties, coordinating the learning process, integrating party results, understanding the characteristics of the training data sets of different participating parties, handling data heterogeneity, and operating with the absence of a verification data set. IBM Federated Learning provides infrastructure and coordination for federated learning. Data scientists can design and run federated learning jobs based on existing, centralized machine learning models and can provide high-level instructions on how to run the federation. The framework applies to both Deep Neural Networks as well as ``traditional'' approaches for the most common machine learning libraries. {\proj} enables data scientists to expand their scope from centralized to federated machine learning, minimizing the learning curve at the outset while also providing the flexibility to deploy to different compute environments and design custom fusion algorithms.

Authors

References32 items

Coordinate-wise Median: Not Bad, Not Bad, Pretty Good

TiFL: A Tier-based Federated Learning System

Advances and Open Problems in Federated Learning

HybridAlpha: An Efficient Approach for Privacy-Preserving Federated Learning

Statistical Model Aggregation via Parameter Matching

Towards Federated Graph Learning for Collaborative Financial Crimes Detection

Bayesian Nonparametric Federated Learning of Neural Networks

BrainTorrent: A Peer-to-Peer Environment for Decentralized Federated Learning

Towards Federated Learning at Scale: System Design

A Hybrid Approach to Privacy-Preserving Federated Learning

A generic framework for privacy preserving deep learning

Zeno: Distributed Stochastic Gradient Descent with Suspicion-based Fault-tolerance

Exploiting Unintended Feature Leakage in Collaborative Learning

Machine Learning with Adversaries: Byzantine Tolerant Gradient Descent

Rényi Differential Privacy

Deep Learning with Differential Privacy

Communication-Efficient Learning of Deep Networks from Decentralized Data

MLlib: Machine Learning in Apache Spark

Scaling Distributed Machine Learning with the Parameter Server

Scikit-learn: Machine Learning in Python

Flask

Practical privacy: the SuLQ framework

A Generalisation, a Simplification and some Applications of Paillier’s Probabilistic Public-Key System

Towards Taming the Resource and Data Heterogeneity in Federated Learning

Google exposed user data, feared repercussions of disclosing to public

Macy’s is warning customers that their information might have been stolen in a data breach, July

Facebook security breach exposes accounts of 50 million users, September 2018

Induction of decision trees

Health Insurance Portability and Accountability Act of 1996.

One such query, q t , requests information about a local model or aggregated information about each party’s data set

The local training process applies a function L that takes query q t and the local dataset D i and outputs a model update r i,t

Yahoo to pay $50 million, offer credit monitoring for massive security breach