ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술지 Gating Mechanism in Deep Neural Networks for Resource-Efficient Continual Learning
Cited 1 time in scopus Download 121 time Share share facebook twitter linkedin kakaostory
진현동, 윤기민, 김은우
IEEE Access, v.10, pp.18776-18786
21HS4800, 장기 시각 메모리 네트워크 기반의 예지형 시각지능 핵심기술 개발, 문진영
Catastrophic forgetting is a well-known tendency in continual learning of a deep neural network to forget previously learned knowledge when optimizing for sequentially incoming tasks. To address the issue, several methods have been proposed in research on continual learning. However, these methods cannot preserve the previously learned knowledge when training for a new task. Moreover, these methods are susceptible to negative interference between tasks, which may lead to catastrophic forgetting. It even becomes increasingly severe when there exists a notable gap between the domains of tasks. This paper proposes a novel method of controlling gates to select a subset of parameters learned for old tasks, which are then used to optimize a new task while avoiding negative interference efficiently. The proposed approach executes the subset of old parameters that provides positive responses by evaluating the effect when the old and new parameters are used together. The execution or skipping of old parameters through the gates is based on several responses across the network. We evaluate the proposed method in different continual learning scenarios involving image classification datasets. The proposed method outperforms other competitive methods and requires fewer parameters than the state-of-the-art methods during inference by applying the proposed gating mechanism that selectively involves a set of old parameters that provides positive prior knowledge to newer tasks. Additionally, we further prove the effectiveness of the proposed method through various analyses.
KSP 제안 키워드
Catastrophic forgetting, Deep neural network(DNN), Image classification, Learning scenarios, Resource-efficient, gating mechanism, novel method, prior knowledge, state-of-The-Art
본 저작물은 크리에이티브 커먼즈 저작자 표시 (CC BY) 조건에 따라 이용할 수 있습니다.
저작자 표시 (CC BY)