Proxy-informed Bayesian transfer learning with unknown sources

TL;DR

This paper introduces PROMPT, a Bayesian transfer learning method that effectively handles unknown source data and negative transfer by using proxy information, applicable even with unobserved task differences and noisy indirect data.

Contribution

It proposes a novel Bayesian approach, PROMPT, that addresses negative transfer without prior source data knowledge, leveraging proxy information to improve transfer learning robustness.

Findings

PROMPT effectively mitigates negative transfer in unknown source scenarios.

Theoretical analysis shows proxy informativeness does not affect negative transfer risk.

PROMPT performs well even with noisy, indirect proxy data.

Abstract

Generalization outside the scope of one's training data requires leveraging prior knowledge about the effects that transfer, and the effects that don't, between different data sources. Transfer learning is a framework for specifying and refining this knowledge about sets of source (training) and target (prediction) data. A challenging open problem is addressing the empirical phenomenon of negative transfer, whereby the transfer learner performs worse on the target data after taking the source data into account than before. We first introduce a Bayesian perspective on negative transfer, and then a method to address it. The key insight from our formulation is that negative transfer can stem from misspecified prior information about non-transferable causes of the source data. Our proposed method, proxy-informed robust method for probabilistic transfer learning (PROMPT), does not require prior knowledge of the source data (the data sources may be "unknown"). PROMPT is thus applicable when differences between tasks are unobserved, such as in the presence of latent confounders. Moreover, the learner need not have access to observations in the target task (may not have the ability to "fine-tune"), and instead makes use of proxy (indirect) information. Our theoretical results show that the threat of negative transfer does not depend on the informativeness of the proxy information, highlighting the usefulness of PROMPT in cases where only noisy indirect information, such as human feedback, is available.