F-MALLOC: Feed-forward Memory Allocation for Continual Learning in Neural Machine Translation
Wu, Junhong, Liu, Yuchen, Zong, Chengqing
–arXiv.org Artificial Intelligence
In the evolving landscape of Neural Machine Translation (NMT), the pretrain-then-finetune paradigm has yielded impressive results. However, the persistent challenge of Catastrophic Forgetting (CF) remains a hurdle. While previous work has introduced Continual Learning (CL) methods to address CF, these approaches grapple with the delicate balance between avoiding forgetting and maintaining system extensibility. To address this, we propose a CL method, named $\textbf{F-MALLOC}$ ($\textbf{F}$eed-forward $\textbf{M}$emory $\textbf{ALLOC}ation)$. F-MALLOC is inspired by recent insights highlighting that feed-forward layers emulate neural memories and encapsulate crucial translation knowledge. It decomposes feed-forward layers into discrete memory cells and allocates these memories to different tasks. By learning to allocate and safeguard these memories, our method effectively alleviates CF while ensuring robust extendability. Besides, we propose a comprehensive assessment protocol for multi-stage CL of NMT systems. Experiments conducted following this new protocol showcase the superior performance of F-MALLOC, evidenced by higher BLEU scores and almost zero forgetting.
arXiv.org Artificial Intelligence
Apr-7-2024
- Country:
- Oceania > Australia
- North America
- Dominican Republic (0.04)
- United States
- Colorado (0.04)
- Washington > King County
- Seattle (0.04)
- Minnesota > Hennepin County
- Minneapolis (0.14)
- California
- San Diego County > San Diego (0.04)
- Los Angeles County > Long Beach (0.04)
- Canada
- Quebec > Montreal (0.04)
- Ontario > Toronto (0.04)
- British Columbia > Metro Vancouver Regional District
- Vancouver (0.14)
- Europe
- Asia
- Japan (0.04)
- Middle East > UAE
- Abu Dhabi Emirate > Abu Dhabi (0.04)
- China
- Africa > Ethiopia
- Addis Ababa > Addis Ababa (0.04)
- Genre:
- Research Report (0.64)
- Technology: