arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1907.09720cs.NEcs.LGstat.ML
收藏

元认知神经记忆

Metalearned Neural Memory

Microsoft Research(微软研究院)

Tsendsuren Munkhdalai, Alessandro Sordoni, Tong Wang, Adam Trischler

摘要:

我们用一种外部记忆机制来增强递归神经网络,这种机制建立在元学习的最新进展之上。我们将这种记忆概念化为一种快速适应的功能,我们将其参数化为一个深层神经网络。从神经记忆函数中读取相当于通过函数推送输入(键向量)以产生输出(值向量)。写入存储器意味着改变功能;具体来说,更新神经网络的参数以编码所需的信息。我们利用元学习的训练和算法技术一次性更新神经记忆功能。所提出的记忆增强模型在各种学习问题上实现了强大的性能,从监督式问答到强化学习。

英文摘要:

We augment recurrent neural networks with an external memory mechanism that builds upon recent progress in metalearning. We conceptualize this memory as a rapidly adaptable function that we parameterize as a deep neural network. Reading from the neural memory function amounts to pushing an input (the key vector) through the function to produce an output (the value vector). Writing to memory means changing the function; specifically, updating the parameters of the neural network to encode desired information. We leverage training and algorithmic techniques from metalearning to update the neural memory function in one shot. The proposed memory-augmented model achieves strong performance on a variety of learning problems, from supervised question answering to reinforcement learning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑