A Reweighted Meta Learning Framework for Robust Few Shot Learning

Model-Agnostic Meta-Learning (MAML) is a popular gradient-based meta-learning framework that tries to find an optimal initialization to minimize the expected loss across all tasks during meta-training. However, it inherently assumes that the contribution of each instance/task to the meta-learner is equal. Therefore, it fails to address the problem of domain differences between base and novel classes in few-shot learning. In this work, we propose a novel and robust meta-learning algorithm, called RW-MAML, which learns to assign weights to training instances or tasks. We consider these weights to be hyper-parameters. Hence, we iteratively optimize the weights using a small set of validation tasks and an online approximation in a \emph{bi-bi-level} optimization framework, in contrast to the standard bi-level optimization in MAML. Therefore, we investigate a practical evaluation setting to demonstrate the scalability of our RW-MAML in two scenarios: (1) out-of-distribution tasks and (2) noisy labels in the meta-training stage. Extensive experiments on synthetic and real-world datasets demonstrate that our framework efficiently mitigates the effects of "unwanted" instances, showing that our proposed technique significantly outperforms state-of-the-art robust meta-learning methods.

[1]  Hugo Larochelle,et al.  Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples , 2019, ICLR.

[2]  Bin Yang,et al.  Learning to Reweight Examples for Robust Deep Learning , 2018, ICML.

[3]  Pengfei Chen,et al.  Understanding and Utilizing Deep Neural Networks Trained with Noisy Labels , 2019, ICML.

[4]  Sergey Levine,et al.  Probabilistic Model-Agnostic Meta-Learning , 2018, NeurIPS.

[5]  Hugo Larochelle,et al.  Optimization as a Model for Few-Shot Learning , 2016, ICLR.

[6]  J. Schulman,et al.  Reptile: a Scalable Metalearning Algorithm , 2018 .

[7]  Alexandros G. Dimakis,et al.  The Robust Manifold Defense: Adversarial Training using Generative Models , 2017, ArXiv.

[8]  Samy Bengio,et al.  Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAML , 2020, ICLR.

[9]  Xian Wu,et al.  Automated Relational Meta-learning , 2020, ICLR.

[10]  Sheng Jin,et al.  Robust Few-Shot Learning for User-Provided Data , 2020, IEEE Transactions on Neural Networks and Learning Systems.

[11]  Sergey Levine,et al.  Meta-Learning with Implicit Gradients , 2019, NeurIPS.

[12]  C A Nelson,et al.  Learning to Learn , 2017, Encyclopedia of Machine Learning and Data Mining.

[13]  Nicolo Fusi,et al.  Weighted Meta-Learning , 2020, ArXiv.

[14]  Hang Li,et al.  Meta-SGD: Learning to Learn Quickly for Few Shot Learning , 2017, ArXiv.

[15]  Richard J. Mammone,et al.  Meta-neural networks that learn by learning , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[16]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[17]  Razvan Pascanu,et al.  Meta-Learning with Latent Embedding Optimization , 2018, ICLR.

[18]  Joshua Achiam,et al.  On First-Order Meta-Learning Algorithms , 2018, ArXiv.

[19]  Roland Vollgraf,et al.  Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms , 2017, ArXiv.

[20]  Sergey Levine,et al.  Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks , 2017, ICML.

[21]  Chun-Nam Yu,et al.  A Direct Approach to Robust Deep Learning Using Adversarial Networks , 2019, ICLR.

[22]  Ananthram Swami,et al.  Distillation as a Defense to Adversarial Perturbations Against Deep Neural Networks , 2015, 2016 IEEE Symposium on Security and Privacy (SP).

[23]  Richard S. Zemel,et al.  Prototypical Networks for Few-shot Learning , 2017, NIPS.

[24]  Amos J. Storkey,et al.  How to train your MAML , 2018, ICLR.

[25]  Qi Zhao,et al.  Foveation-based Mechanisms Alleviate Adversarial Examples , 2015, ArXiv.

[26]  Andrew Y. Ng,et al.  Reading Digits in Natural Images with Unsupervised Feature Learning , 2011 .

[27]  Oriol Vinyals,et al.  Matching Networks for One Shot Learning , 2016, NIPS.

[28]  Daan Wierstra,et al.  Meta-Learning with Memory-Augmented Neural Networks , 2016, ICML.

[29]  Latifur Khan,et al.  A Primal-Dual Subgradient Approach for Fair Meta Learning , 2020, 2020 IEEE International Conference on Data Mining (ICDM).

[30]  Philip H. S. Torr,et al.  Alpha MAML: Adaptive Model-Agnostic Meta-Learning , 2019, ArXiv.

[31]  Eunho Yang,et al.  Learning to Balance: Bayesian Meta-Learning for Imbalanced and Out-of-distribution Tasks , 2019, ICLR.

[32]  Qi Xie,et al.  Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting , 2019, NeurIPS.

[33]  Sergey Levine,et al.  Online Meta-Learning , 2019, ICML.

[34]  Joseph J. Lim,et al.  Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation , 2019, NeurIPS.

[35]  Xingrui Yu,et al.  Co-teaching: Robust training of deep neural networks with extremely noisy labels , 2018, NeurIPS.