Conditional Meta-Learning of Linear Representations
This addresses the problem of representation learning in multi-task environments for machine learning practitioners, offering a conditional approach that is incremental over existing meta-learning methods.
The paper tackles the limitation of standard meta-learning in capturing task nuances by inferring a conditioning function that maps task side information into tailored representations, showing outperformance in clustered task environments and proposing a meta-algorithm with faster learning rates and fewer hyper-parameters in unconditional settings.
Standard meta-learning for representation learning aims to find a common representation to be shared across multiple tasks. The effectiveness of these methods is often limited when the nuances of the tasks' distribution cannot be captured by a single representation. In this work we overcome this issue by inferring a conditioning function, mapping the tasks' side information (such as the tasks' training dataset itself) into a representation tailored to the task at hand. We study environments in which our conditional strategy outperforms standard meta-learning, such as those in which tasks can be organized in separate clusters according to the representation they share. We then propose a meta-algorithm capable of leveraging this advantage in practice. In the unconditional setting, our method yields a new estimator enjoying faster learning rates and requiring less hyper-parameters to tune than current state-of-the-art methods. Our results are supported by preliminary experiments.