Regularising Knowledge Transfer by Meta Functional Learning

Machine learning classifiers’ capability is largely dependent on the scale of available training data and limited by the model overfitting in data-scarce learning tasks. To address this problem, this work proposes a novel Meta Functional Learning (MFL) by meta-learning a generalisable functional model from data-rich tasks whilst simultaneously regularising knowledge transfer to data-scarce tasks. The MFL computes meta-knowledge on functional regularisation generalisable to different learning tasks by which functional training on limited labelled data promotes more discriminative functions to be learned. Moreover, we adopt an Iterative Update strategy on MFL (MFL-IU). This improves knowledge transfer regularisation from MFL by progressively learning the functional regularisation in knowledge transfer. Experiments on three Few-Shot Learning (FSL) benchmarks (miniImageNet, CIFAR-FS and CUB) show that meta functional learning for regularisation knowledge transfer can benefit improving FSL classifiers.

Temporal expression divergence of network modules

Here we propose new module-based approaches to identify differentially regulated network sub-modules combining temporal trajectories of expression profiles with static network skeletons. Starting from modules identified by network clustering of static networks, our analysis refines pre-defined genesets by partitioning them into smaller homogeneous sets by non-paramettric Bayesian methods. Especially for case-control time series data we developed multi-time point discriminative models and identified each network module as a mixture or admixture of dynamic discriminative functions. Our results shows that our proposed approach outperformed existing geneset enrichment methods in simulation studies. Moreover we applied the methods to neural stem cell differentiation data, and discovered novel modules differentially perturbed in different developmental stages.