Timezone: »

Learning to Rehearse in Long Sequence Memorization
Zhu Zhang · Chang Zhou · Jianxin Ma · Zhijie Lin · Jingren Zhou · Hongxia Yang · Zhou Zhao

Thu Jul 22 09:00 AM -- 11:00 AM (PDT) @ None #None

Existing reasoning tasks often have an important assumption that the input contents can be always accessed while reasoning, requiring unlimited storage resources and suffering from severe time delay on long sequences. To achieve efficient reasoning on long sequences with limited storage resources, memory augmented neural networks introduce a human-like write-read memory to compress and memorize the long input sequence in one pass, trying to answer subsequent queries only based on the memory. But they have two serious drawbacks: 1) they continually update the memory from current information and inevitably forget the early contents; 2) they do not distinguish what information is important and treat all contents equally. In this paper, we propose the Rehearsal Memory (RM) to enhance long-sequence memorization by self-supervised rehearsal with a history sampler. To alleviate the gradual forgetting of early information, we design self-supervised rehearsal training with recollection and familiarity tasks. Further, we design a history sampler to select informative fragments for rehearsal training, making the memory focus on the crucial information. We evaluate the performance of our rehearsal memory by the synthetic bAbI task and several downstream tasks, including text/video question answering and recommendation on long sequences.

Author Information

Zhu Zhang (DAMO Academy, Alibaba Group,)
Chang Zhou (Alibaba Group)
Jianxin Ma (Alibaba Group)
Zhijie Lin (Zhejiang University)
Jingren Zhou (Alibaba Group)
Hongxia Yang (Alibaba Group)
Zhou Zhao (Zhejiang University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors