Can I read The Effect of Diversity in Meta-Learning on EtoBox?

The Effect of Diversity in Meta-Learning by Kumar, Ramnath; Deleu, Tristan; Bengio, Yoshua is a scholarly article available to read on EtoBox.

What is The Effect of Diversity in Meta-Learning about?

Recent studies show that task distribution plays a vital role in the meta-learner's performance. Conventional wisdom is that task diversity should improve the performance of meta-learning. In this work, we find evidence to the contrary; (i) our experiments draw into question the efficacy of our learned models: similar manifolds can be learned with a subset of the data (lower task diversity). This finding questions the advantage of providing more data to the model, and (ii) adding diversity to the task distribution (higher task diversity) sometimes hinders the model and does not lead to a significant improvement in performance as previously believed. To strengthen our findings, we provide both empirical and theoretical evidence.

Author

Kumar, Ramnath; Deleu, Tristan; Bengio, Yoshua

Published

2022

Language

EN