paper-with-me

홈 › Papers

Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning

2021-03-16 · ICLR 2021 1 · Namyeong Kwon, Hwidong Na, Gabriel Huang, Simon Lacoste-Julien

Model-agnostic meta-learning (MAML) is a popular method for few-shot learning but assumes that we have access to the meta-training set. In practice, training on the meta-training set may not always be an option due to data privacy concerns, intellectual property issues, or merely lack of computing resources. In this paper, we consider the novel problem of repurposing pretrained MAML checkpoints to solve new few-shot classification tasks. Because of the potential distribution mismatch, the original MAML steps may no longer be optimal. Therefore we propose an alternative meta-testing procedure and combine MAML gradient steps with adversarial training and uncertainty-based stepsize adaptation. Our method outperforms "vanilla" MAML on same-domain and cross-domains benchmarks using both SGD and Adam optimizers and shows improved robustness to the choice of base stepsize.

📄 PDF Abstract BibTeX arXiv:2103.09027

Code (1)

NamyeongK/USA_UFGSM 공식 구현 pytorch

Tasks

Few-Shot LearningMeta-Learning

Methods 이 논문이 사용한 방법론

SGD Stochastic Gradient Descent is an iterative optimization technique that uses minibatches of data to form an expectation of the gradient, rather than the full gradient using…
MAML 설명 없음
Adam 설명 없음

Similar Papers 제목 키워드 기반

Supervised Classification Heads as Semantic Prototypes: Unlocking Vision-Language Alignment via Weight Recycling

2026-05-21 · David Méndez, Roberto Confalonieri, Natalia Díaz Rodríguez arxiv

Vision-Language Models (VLMs) excel at tasks like zero-shot classification and cross-modal retrieval by mapping images and text to a shared space, but this requires expensive end-to-end training with massive paired datas…

Cross-Modal RetrievalData Augmentation

Domain Expansion of Image Generators

2023-01-12 · CVPR 2023 1 · Yotam Nitzan, Michaël Gharbi, Richard Zhang, Taesung Park 외

Can one inject new concepts into an already trained generative model, while respecting its existing structure and knowledge? We propose a new task - domain expansion - to address this. Given a pretrained generator and no…

Keypoint Action Tokens Enable In-Context Imitation Learning in Robotics

2024-03-28 · Norman Di Palo, Edward Johns

We show that off-the-shelf text-based Transformers, with no additional training, can perform few-shot in-context visual imitation learning, mapping visual observations to action sequences that emulate the demonstrator's …

Imitation Learning

Distilling Internet-Scale Vision-Language Models into Embodied Agents

2023-01-29 · Theodore Sumers, Kenneth Marino, Arun Ahuja, Rob Fergus 외

Instruction-following agents must ground language into their observation and action spaces. Learning to ground language is challenging, typically requiring domain-specific engineering or large quantities of human interac…

Instruction Following

Few-shot link prediction via graph neural networks for Covid-19 drug-repurposing

2020-07-20 · Vassilis N. Ioannidis, Da Zheng, George Karypis

Predicting interactions among heterogenous graph structured data has numerous applications such as knowledge graph completion, recommendation systems and drug discovery. Often times, the links to be predicted belong to r…

Drug DiscoveryFew-Shot LearningGraph LearningKnowledge Graph Completion+3