arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.05922cs.LGcs.AIstat.ML
收藏

用于流媒体学习的内存高效体验回放

Memory Efficient Experience Replay for Streaming Learning

Carlson Center for Imaging Science(卡尔森成像科学中心);Rochester Institute of Technology(罗切斯特理工学院)

Tyler L. Hayes, Nathan D. Cahill, Christopher Kanan

摘要:

在监督式机器学习中,代理通常被训练一次,然后部署。虽然这在静态环境中很好,但机器人通常在不断变化的环境中运行,必须从数据流中快速学习新事物。在这种被称为流学习的范例中,学习者是从一个不能被假设为独立同分布(iid)的数据流中在线训练的。流学习会导致传统的深度神经网络(DNN)失败,原因有两个:1)它们需要多次通过整个数据集; 2)非iid数据会导致灾难性遗忘。解决这两个问题的一个老办法是排练。为了学习一个新的例子,排练将它与以前的例子混合,然后使用这种混合来更新DNN。完全排练是缓慢和记忆密集型的,因为它存储了所有先前观察到的例子,并且其防止灾难性遗忘的有效性尚未在现代DNN中研究。在这里,我们描述了ExStream算法的内存高效排练,并比较它的替代品。我们发现,完全排练可以消除各种流式学习设置中的灾难性遗忘,并且ExStream使用更少的内存和计算量就能表现良好。

英文摘要:

In supervised machine learning, an agent is typically trained once and then deployed. While this works well for static settings, robots often operate in changing environments and must quickly learn new things from data streams. In this paradigm, known as streaming learning, a learner is trained online, in a single pass, from a data stream that cannot be assumed to be independent and identically distributed (iid). Streaming learning will cause conventional deep neural networks (DNNs) to fail for two reasons: 1) they need multiple passes through the entire dataset; and 2) non-iid data will cause catastrophic forgetting. An old fix to both of these issues is rehearsal. To learn a new example, rehearsal mixes it with previous examples, and then this mixture is used to update the DNN. Full rehearsal is slow and memory intensive because it stores all previously observed examples, and its effectiveness for preventing catastrophic forgetting has not been studied in modern DNNs. Here, we describe the ExStream algorithm for memory efficient rehearsal and compare it to alternatives. We find that full rehearsal can eliminate catastrophic forgetting in a variety of streaming learning settings, with ExStream performing well using far less memory and computation.

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。

↑