Provable Adaptation across Multiway Domains via Representation Learning
arXiv:2106.06657
Abstract
This paper studies zero-shot domain adaptation where each domain is indexed on a multi-dimensional array, and we only have data from a small subset of domains. Our goal is to produce predictors that perform well on \emph{unseen} domains. We propose a model which consists of a domain-invariant latent representation layer and a domain-specific linear prediction layer with a low-rank tensor structure. Theoretically, we present explicit sample complexity bounds to characterize the prediction error on unseen domains in terms of the number of domains with training data and the number of data per domain. To our knowledge, this is the first finite-sample guarantee for zero-shot domain adaptation. In addition, we provide experiments on two-way MNIST and four-way fiber sensing datasets to demonstrate the effectiveness of our proposed model.
References in corpus (9)
- A Simpler Approach to Matrix Completion
- Deep Multi-task Representation Learning: A Tensor Factorisation Approach
- The Benefit of Multitask Representation Learning
- A Unified Perspective on Multi-Domain and Multi-Task Learning
- Feature-Critic Networks for Heterogeneous Domain Generalization
- On the Theory of Transfer Learning: The Importance of Task Diversity
- Few-Shot Learning via Learning the Representation, Provably
- Learning Tensors in Reproducing Kernel Hilbert Spaces with Multilinear Spectral Penalties
- Tensor Completion Made Practical