Multimodal Model-Agnostic Meta-Learning via Task-Aware Modulation

Abstract

Model-agnostic meta-learners aim to acquire meta-learned parameters from similar tasks to adapt to novel tasks from the same distribution with few gradient updates. With the flexibility in the choice of models, those frameworks demonstrate appealing performance on a variety of domains such as few-shot image classification and reinforcement learning. However, one important limitation of such frameworks is that they seek a common initialization shared across the entire task distribution, substantially limiting the diversity of the task distributions that they are able to learn from. In this paper, we augment MAML with the capability to identify the mode of tasks sampled from a multimodal task distribution and adapt quickly through gradient updates. Specifically, we propose a multimodal MAML (MMAML) framework, which is able to modulate its meta-learned prior parameters according to the identified mode, allowing more efficient fast adaptation. We evaluate the proposed model on a diverse set of few-shot learning tasks, including regression, image classification, and reinforcement learning. The results not only demonstrate the effectiveness of our model in modulating the meta-learned prior in response to the characteristics of tasks but also show that training on a multimodal distribution can produce an improvement over unimodal training.

Cited by in corpus (23)

• Meta-learning in healthcare: A survey
• How to Train Your MAML to Excel in Few-Shot Classification
• Learning Adaptive Classifiers Synthesis for Generalized Few-Shot Learning
• Improving Generalization in Meta-learning via Task Augmentation
• Learning to Synthesize Programs as Interpretable and Generalizable Policies
• Online Structured Meta-learning
• Structured Prediction for Conditional Meta-Learning
• A Channel Coding Benchmark for Meta-Learning
• Task-adaptive Neural Process for User Cold-Start Recommendation
• The Advantage of Conditional Meta-Learning for Biased Regularization and Fine-Tuning
• HetMAML: Task-Heterogeneous Model-Agnostic Meta-Learning for Few-Shot Learning Across Modalities
• Meta-KD: A Meta Knowledge Distillation Framework for Language Model Compression across Domains
• Off-Policy Meta-Reinforcement Learning Based on Feature Embedding Spaces
• ST-MAML: A Stochastic-Task based Method for Task-Heterogeneous Meta-Learning
• An Optimization-Based Meta-Learning Model for MRI Reconstruction with Diverse Dataset
• Reinforced Few-Shot Acquisition Function Learning for Bayesian Optimization
• Model-Agnostic Learning to Meta-Learn
• Combining Domain-Specific Meta-Learners in the Parameter Space for Cross-Domain Few-Shot Classification
• Meta ordinal weighting net for improving lung nodule classification
• Covariate Distribution Aware Meta-learning
• Towards Enabling Meta-Learning from Target Models
• Relation-aware Meta-learning for Market Segment Demand Prediction with Limited Records
• Self-Denoising Neural Networks for Few Shot Learning