Network Transplanting (extended abstract)
This addresses the bottleneck of needing to collect all data upfront for learning massive tasks and categories, enabling incremental addition of new categories without affecting existing ones.
The paper tackles the problem of transplanting a category-specific neural network module into a generic modular network without requiring extensive labeled data, achieving performance that surpasses a baseline using 100 training samples even when using no training samples.
This paper focuses on a new task, i.e., transplanting a category-and-task-specific neural network to a generic, modular network without strong supervision. We design a functionally interpretable structure for the generic network. Like building LEGO blocks, we teach the generic network a new category by directly transplanting the module corresponding to the category from a pre-trained network with a few or even without sample annotations. Our method incrementally adds new categories to the generic network but does not affect representations of existing categories. In this way, our method breaks the typical bottleneck of learning a net for massive tasks and categories, i.e., the requirement of collecting samples for all tasks and categories at the same time before the learning begins. Thus, we use a new distillation algorithm, namely back-distillation, to overcome specific challenges of network transplanting. Our method without training samples even outperformed the baseline with 100 training samples.