Large-Scale Meta-Learning with Continual Trajectory Shifting

2021-02-14 · Jaewoong Shin, Hae Beom Lee, Boqing Gong, Sung Ju Hwang

Meta-learning of shared initialization parameters has shown to be highly effective in solving few-shot learning tasks. However, extending the framework to many-shot scenarios, which may further enhance its practicality, has been relatively overlooked due to the technical difficulties of meta-learning over long chains of inner-gradient steps. In this paper, we first show that allowing the meta-learners to take a larger number of inner gradient steps better captures the structure of heterogeneous and large-scale task distributions, thus results in obtaining better initialization points. Further, in order to increase the frequency of meta-updates even with the excessively long inner-optimization trajectories, we propose to estimate the required shift of the task-specific parameters with respect to the change of the initialization parameters. By doing so, we can arbitrarily increase the frequency of meta-updates and thus greatly improve the meta-level convergence as well as the quality of the learned initializations. We validate our method on a heterogeneous set of large-scale tasks and show that the algorithm largely outperforms the previous first-order meta-learning methods in terms of both generalization performance and convergence, as well as multi-task learning and fine-tuning baselines.

📄 PDF Abstract BibTeX arXiv:2102.07215

Code (0)

등록된 구현이 없습니다.

Tasks

Few-Shot Learning Meta-Learning Multi-Task Learning

On the Stability-Plasticity Dilemma in Continual Meta-Learning: Theory and Algorithm

2023-09-21 · NeurIPS 2023 11

We focus on Continual Meta-Learning (CML), which targets accumulating and exploiting meta-knowledge on a sequence of non-i.i.d. tasks. The primary challenge is to strike a balance between stability and plasticity, where …

Learning to Adapt to Online Streams with Distribution Shifts

2023-03-02 · Chenyan Wu, Yimu Pan, Yandong Li, James Z. Wang

Test-time adaptation (TTA) is a technique used to reduce distribution gaps between the training and testing sets by leveraging unlabeled test data during inference. In this work, we expand TTA to a more practical scenari…

Benchmarking Meta-Learning Test-time Adaptation Video Segmentation+1

Continual Meta-Reinforcement Learning for UAV-Aided Vehicular Wireless Networks

2022-07-13 · Riccardo Marini, Sangwoo Park, Osvaldo Simeone, Chiara Buratti

Unmanned aerial base stations (UABSs) can be deployed in vehicular wireless networks to support applications such as extended sensing via vehicle-to-everything (V2X) services. A key problem in such systems is designing a…

Meta Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)+1

MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wild

2026-03-17 · Peng Xia, Jianwen Chen, Xinyu Yang, Haoqin Tu 외 arxiv

Large language model (LLM) agents are increasingly used for complex tasks, yet deployed agents often remain static, failing to adapt as user needs evolve. This creates a tension between the need for continuous service an…

Reinforcement Learning

KOPPA: Improving Prompt-based Continual Learning with Key-Query Orthogonal Projection and Prototype-based One-Versus-All

2023-11-26 · Quyen Tran, Hoang Phan, Lam Tran, Khoat Than 외

Drawing inspiration from prompt tuning techniques applied to Large Language Models, recent methods based on pre-trained ViT networks have achieved remarkable results in the field of Continual Learning. Specifically, thes…

All Continual Learning Meta-Learning