Skip to yearly menu bar Skip to main content


Poster

Context-Aware Meta-Learning

Christopher Fifty · Dennis Duan · Ronald Junkins · Ehsan Amid · Jure Leskovec · Christopher Re · Sebastian Thrun

Halle B #178
[ ] [ Project Page ]
Thu 9 May 7:30 a.m. PDT — 9:30 a.m. PDT

Abstract:

Large Language Models like ChatGPT demonstrate a remarkable capacity to learn new concepts during inference without any fine-tuning. However, visual models trained to detect new objects during inference have been unable to replicate this ability, and instead either perform poorly or require meta-training and/or fine-tuning on similar objects. In this work, we propose a meta-learning algorithm that emulates Large Language Models by learning new visual concepts during inference without fine-tuning. Our approach leverages a frozen pre-trained feature extractor, and analogous to in-context learning, recasts meta-learning as sequence modeling over datapoints with known labels and a test datapoint with an unknown label. On 8 out of 11 meta-learning benchmarks, our approach---without meta-training or fine-tuning---exceeds or matches the state-of-the-art algorithm, P>M>F, which is meta-trained on these benchmarks.

Chat is not available.