Timezone: »

Fast and Flexible Multi-Task Classification using Conditional Neural Adaptive Processes
James Requeima · Jonathan Gordon · John Bronskill · Sebastian Nowozin · Richard Turner

Wed Dec 11 04:05 PM -- 04:10 PM (PST) @ West Ballroom A + B

The goal of this paper is to design image classification systems that, after an initial multi-task training phase, can automatically adapt to new tasks encountered at test time. We introduce a conditional neural process based approach to the multi-task classification setting for this purpose, and establish connections to the meta- and few-shot learning literature. The resulting approach, called CNAPs, comprises a classifier whose parameters are modulated by an adaptation network that takes the current task's dataset as input. We demonstrate that CNAPs achieves state-of-the-art results on the challenging Meta-Dataset benchmark indicating high-quality transfer-learning. We show that the approach is robust, avoiding both over-fitting in low-shot regimes and under-fitting in high-shot regimes. Timing experiments reveal that CNAPs is computationally efficient at test-time as it does not involve gradient based adaptation. Finally, we show that trained models are immediately deployable to continual learning and active learning where they can outperform existing approaches that do not leverage transfer learning.

Author Information

James Requeima (University of Cambridge / Invenia Labs)
Jonathan Gordon (University of Cambridge)
John Bronskill (University of Cambridge)
Sebastian Nowozin (Google Research Berlin)
Richard Turner (Cambridge)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors