Game-theoretic distributed learning of generative models for heterogeneous data collections
This addresses the problem of handling heterogeneous data in distributed learning for AI/ML applications, offering a novel method but with incremental improvements over existing generative model techniques.
The paper tackles the challenge of distributed learning with heterogeneous local models and data by proposing a game-theoretic approach that exchanges synthetic data instead of model parameters, proving convergence to a unique Nash equilibrium for exponential family models and demonstrating advantages on standard vision datasets for image classification and conditional generation.
One of the main challenges in distributed learning arises from the difficulty of handling heterogeneous local models and data. In light of the recent success of generative models, we propose to meet this challenge by building on the idea of exchanging synthetic data instead of sharing model parameters. Local models can then be treated as ``black boxes'' with the ability to learn their parameters from data and to generate data according to these parameters. Moreover, if the local models admit semi-supervised learning, we can extend the approach by enabling local models on different probability spaces. This allows to handle heterogeneous data with different modalities. We formulate the learning of the local models as a cooperative game starting from the principles of game theory. We prove the existence of a unique Nash equilibrium for exponential family local models and show that the proposed learning approach converges to this equilibrium. We demonstrate the advantages of our approach on standard benchmark vision datasets for image classification and conditional generation.