Foundations of Bayesian Learning from Synthetic Data
This addresses the challenge of ensuring reliable machine learning in privacy-sensitive or data-scarce environments, though it is incremental as it builds on existing Bayesian methods.
The paper tackles the problem of learning from synthetic data, especially when augmenting real data with synthetic data from another party, by using a Bayesian paradigm to characterize model parameter updating, and demonstrates that a novel decision-theoretic approach outperforms standard methods in supervised learning and inference experiments.
There is significant growth and interest in the use of synthetic data as an enabler for machine learning in environments where the release of real data is restricted due to privacy or availability constraints. Despite a large number of methods for synthetic data generation, there are comparatively few results on the statistical properties of models learnt on synthetic data, and fewer still for situations where a researcher wishes to augment real data with another party's synthesised data. We use a Bayesian paradigm to characterise the updating of model parameters when learning in these settings, demonstrating that caution should be taken when applying conventional learning algorithms without appropriate consideration of the synthetic data generating process and learning task. Recent results from general Bayesian updating support a novel and robust approach to Bayesian synthetic-learning founded on decision theory that outperforms standard approaches across repeated experiments on supervised learning and inference problems.