Recurrent Neural Networks

Simple Neural Attention Meta-Learner

Introduced by Mishra et al. in A Simple Neural Attentive Meta-Learner

The Simple Neural Attention Meta-Learner, or SNAIL, combines the benefits of temporal convolutions and attention to solve meta-learning tasks. They introduce positional dependence through temporal convolutions to make the model applicable to reinforcement tasks - where the observations, actions, and rewards are intrinsically sequential. They also introduce attention in order to provide pinpoint access over an infinitely large context. SNAIL is constructing by combining the two: we use temporal convolutions to produce the context over which we use a causal attention operation.

Source: A Simple Neural Attentive Meta-Learner

Papers


Paper Code Results Date Stars

Tasks


Task Papers Share
Meta-Learning 2 18.18%
Navigate 1 9.09%
Anomaly Detection 1 9.09%
Image Segmentation 1 9.09%
Semantic Segmentation 1 9.09%
Style Transfer 1 9.09%
Few-Shot Learning 1 9.09%
Metric Learning 1 9.09%
Sequential skip prediction 1 9.09%

Categories