Robust Domain Generalisation by Enforcing Distribution Invariance / 1455
Sarah M. Erfani, Mahsa Baktashmotlagh, Masud Moshtaghi, Vinh Nguyen, Christopher Leckie, James Bailey, Kotagiri Ramamohanarao
Many conventional statistical machine learning algorithms generalise poorly if distribution bias exists in the datasets. For example, distribution bias arises in the context of domain generalization, where knowledge acquired from multiple source domains need to be used in a previously unseen target domains. We propose Elliptical Summary Randomisation (ESRand), an efficient domain generalisation approach that comprises of a randomized kernel and elliptical data summarisation. ESRand learns a domain interdependent projection to a latent subspace that minimises the existing biases to the data while maintaining the functional relationship between domains. In the latent subspace, ellipsoidal summaries replace the samples to enhance the generalisation by further removing biasand noise in the data. Moreover, the summarization enables large-scale data processing by significantly reducing the size of the data. Through comprehensive analysis, we show that our subspace-based approach outperforms state-of-the-art results on several activity recognition benchmark datasets, while keeping the computational complexity significantly low.