• Corpus ID: 235358966

Meta-Learning with Fewer Tasks through Task Interpolation

  title={Meta-Learning with Fewer Tasks through Task Interpolation},
  author={Huaxiu Yao and Linjun Zhang and Chelsea Finn},
Meta-learning enables algorithms to quickly learn a newly encountered task with just a few labeled examples by transferring previously learned knowledge. However, the bottleneck of current meta-learning algorithms is the requirement of a large number of meta-training tasks, which may not be accessible in real-world scenarios. To address the challenge that available tasks may not densely sample the space of tasks, we propose to augment the task set through interpolation. By meta-learning with… 

Adversarial Training Helps Transfer Learning via Better Representations

It is demonstrated both theoretically and empirically that semi-supervised learning in the source data can also improve transfer learning by similarly improving the representation, and that performing adversarial training on top of semi- supervised learning can further improve transferability, suggesting that the two approaches have complementary benefits on representations.

Regularization Guarantees Generalization in Bayesian Reinforcement Learning through Algorithmic Stability

This work shows that regularized MDPs satisfy a certain quadratic growth criterion, which is sufficient to establish stability, and allows us to study the effect of regularization on generalization in the Bayesian RL setting.

Exploiting Style Transfer-based Task Augmentation for Cross-Domain Few-Shot Learning

The proposed Task Augmented Meta-Learning (TAML) increases the diversity of styles of training tasks, and contributes to training a model with better domain generalization ability.

Set-based Meta-Interpolation for Few-Task Meta-Learning

A novel domain-agnostic task augmentation method, Meta-Interpolation, which utilizes expressive neural set functions to densify the meta-training task distribution using bilevel optimization is proposed and it is proved that task interpolation with the set function regularizes theMeta-learner to improve generalization.

Interval Bound Propagation-aided Few-shot Learning

This work introduces the notion of interval bounds from the provably robust training literature to few-shot learning and introduces a novel strategy to artificially form new tasks for training by interpolating between the available tasks and their respective interval bounds to aid in cases with a scarcity of tasks.

Learning to generate imaginary tasks for improving generalization in meta-learning

This paper seeks an approach that up-samples meta-training tasks from the task representation via a task up-sampling network and suffices to generate tasks that can maximally contribute to the latest meta-learner by maximizing an adversarial loss.

C-Mixup: Improving Generalization in Regression

A simple yet powerful algorithm, C-Mixup, to improve generalization on regression tasks by selectively interpolating examples with similar labels that mitigates the effects of domain-associated information and yields domain-invariant representations.

Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts

This work proposes a novel framework for unsupervised test-time adaptation, which is formulated as a knowledge distillation process to address domain shift and incorporates Mixture-of-Experts (MoE) as teachers, where each expert is separately trained on different source domains to maximize their speciality.

Improving Meta-Learning Generalization with Activation-Based Early-Stopping

Activation Based Early-st stopping (ABE) is proposed, an alternative to using validation-based early-stopping for meta-learning, and results show that the method validation- based early-Stopping, for all three algorithms, can vary significantly.

Meta Reinforcement Learning with Finite Training Tasks - a Density Estimation Approach

This work directly learn the task distribution, using density estimation techniques, and then train a policy on the learned task distribution and demonstrates that this regularization implied by the kernel density estimation method is useful in practice, when ‘plugged in’ the state-of-the-art VariBAD meta RL algorithm.



Improving Generalization in Meta-learning via Task Augmentation

Two task augmentation methods are proposed, including MetaMix and Channel Shuffle, which outperform state-of-the-art results by a large margin across many datasets and are compatible with existing meta-learning algorithms.

Task Agnostic Meta-Learning for Few-Shot Learning

An entropy-based approach that meta-learns an unbiased initial model with the largest uncertainty over the output labels by preventing it from over-performing in classification tasks, which outperforms compared meta-learning algorithms in both few-shot classification and reinforcement learning tasks.

Concept Learners for Few-Shot Learning

Developing algorithms that are able to generalize to a novel task given only a few labeled examples represents a fundamental challenge in closing the gap between machine- and human-level performance.

Probabilistic Model-Agnostic Meta-Learning

This paper proposes a probabilistic meta-learning algorithm that can sample models for a new task from a model distribution that is trained via a variational lower bound, and shows how reasoning about ambiguity can also be used for downstream active learning problems.

Regularizing Meta-Learning via Gradient Dropout

This paper introduces a simple yet effective method to alleviate the risk of overfitting for gradient-based meta-learning by randomly dropping the gradient in the inner-loop optimization of each parameter in deep neural networks, such that the augmented gradients improve generalization to new tasks.

Provable Meta-Learning of Linear Representations

This paper provides provably fast, sample-efficient algorithms to address the dual challenges of learning a common set of features from multiple, related tasks and transferring this knowledge to new, unseen tasks, which are central to the general problem of meta-learning.

DReCa: A General Task Augmentation Strategy for Few-Shot Natural Language Inference

DReCA (Decomposing datasets into Reasoning Categories), a simple method for discovering and using latent reasoning categories in a dataset, to form additional high quality tasks to improve the accuracy of meta-learners.

Gradient-Based Meta-Learning with Learned Layerwise Metric and Subspace

This work demonstrates that the dimension of this learned subspace reflects the complexity of the task-specific learner's adaptation task, and also that the model is less sensitive to the choice of initial learning rates than previous gradient-based meta-learning methods.

A Simple Neural Attentive Meta-Learner

This work proposes a class of simple and generic meta-learner architectures that use a novel combination of temporal convolutions and soft attention; the former to aggregate information from past experience and the latter to pinpoint specific pieces of information.

Meta-Learning Requires Meta-Augmentation

It is demonstrated that meta-augmentation produces large complementary benefits to recently proposed meta-regularization techniques, and is described as a way to add randomness that discourages the base learner and model from learning trivial solutions that do not generalize to new tasks.