Alpha MAML: Adaptive Model-Agnostic Meta-Learning
This work addresses a practical issue for researchers and practitioners using MAML in few-shot learning, though it is incremental as it builds directly on the existing MAML framework.
The paper tackles the problem of costly hyperparameter tuning in Model-Agnostic Meta-Learning (MAML) by introducing Alpha MAML, an extension that incorporates an online hyperparameter adaptation scheme, resulting in a substantial reduction in tuning needs and improved training stability with less sensitivity to hyperparameter choice, as demonstrated on the Omniglot database.
Model-agnostic meta-learning (MAML) is a meta-learning technique to train a model on a multitude of learning tasks in a way that primes the model for few-shot learning of new tasks. The MAML algorithm performs well on few-shot learning problems in classification, regression, and fine-tuning of policy gradients in reinforcement learning, but comes with the need for costly hyperparameter tuning for training stability. We address this shortcoming by introducing an extension to MAML, called Alpha MAML, to incorporate an online hyperparameter adaptation scheme that eliminates the need to tune meta-learning and learning rates. Our results with the Omniglot database demonstrate a substantial reduction in the need to tune MAML training hyperparameters and improvement to training stability with less sensitivity to hyperparameter choice.