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 1701.08718cs.LGcs.NEstat.ML
收藏

具有虫洞连接的记忆增强神经网络

Memory Augmented Neural Networks with Wormhole Connections

Montreal Institute for Learning Algorithms(蒙特利尔学习算法研究所);Universite de Montreal(蒙特利尔大学)

Caglar Gulcehre, Sarath Chandar, Yoshua Bengio

摘要:

最近关于长期依赖任务的实证结果表明,与香草递归神经网络(RNN)相比,增强了外部记忆的神经网络可以更容易地学习长期依赖任务,并实现更好的泛化。我们认为,记忆增强的神经网络可以通过创建捷径(或虫洞)连接来减少消失梯度的影响。基于这一观察,我们提出了一种新的记忆增强神经网络模型称为TARDIS(时序自动关系发现序列)。TARDIS的控制器可以将自己先前隐藏状态的一组选择性嵌入存储到外部存储器中,并在需要时重新访问它们。对于TARDIS,记忆充当虫洞连接到过去的存储器,以更有效地传播梯度,并有助于学习时间依赖性。TARDIS的内存结构与神经图灵机(NTM)和动态神经图灵机(D-NTM)都有相似之处,但TARDIS的读写操作都更简单,更有效。我们使用离散寻址的读/写操作,这有助于大大减少消失的梯度问题,非常长的序列。一旦内存变满,TARDIS中的读写操作就与启发式算法联系在一起,这使得学习问题比NTM或D-NTM类型的架构更简单。我们提供了一个详细的分析梯度传播一般的人。我们在不同的长期依赖性任务上评估我们的模型,并报告所有这些任务的竞争结果。

英文摘要:

Recent empirical results on long-term dependency tasks have shown that neural networks augmented with an external memory can learn the long-term dependency tasks more easily and achieve better generalization than vanilla recurrent neural networks (RNN). We suggest that memory augmented neural networks can reduce the effects of vanishing gradients by creating shortcut (or wormhole) connections. Based on this observation, we propose a novel memory augmented neural network model called TARDIS (Temporal Automatic Relation Discovery in Sequences). The controller of TARDIS can store a selective set of embeddings of its own previous hidden states into an external memory and revisit them as and when needed. For TARDIS, memory acts as a storage for wormhole connections to the past to propagate the gradients more effectively and it helps to learn the temporal dependencies. The memory structure of TARDIS has similarities to both Neural Turing Machines (NTM) and Dynamic Neural Turing Machines (D-NTM), but both read and write operations of TARDIS are simpler and more efficient. We use discrete addressing for read/write operations which helps to substantially to reduce the vanishing gradient problem with very long sequences. Read and write operations in TARDIS are tied with a heuristic once the memory becomes full, and this makes the learning problem simpler when compared to NTM or D-NTM type of architectures. We provide a detailed analysis on the gradient propagation in general for MANNs. We evaluate our models on different long-term dependency tasks and report competitive results in all of them.

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。

↑