Abstract: We propose to address the problem of few-shot classification by meta-learning "what to observe" and "where to attend" in a relational perspective. Our method lever-ages relational patterns ...
Abstract: The previous relation-based knowledge distillation methods tend to construct global similarity relationship matrix in a mini-batch while ignoring the knowledge of neighbourhood relationship.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback