知行编程网知行编程网  2022-05-19 15:00 知行编程网 隐藏边栏 |   抢沙发  3 
文章评分 0 次,平均分 0.0

有了Transformer框架后是不是RNN完全可以废弃了?

来自 | 知乎   作者 | 吴俣
https://www.zhihu.com/question/302392659/answer/551542493
编辑 | 机器学习算法与自然语言处理
本文仅作学术分享,如有侵权,请联系删除

有了Transformer框架后是不是RNN完全可以废弃了?

我们首先区分一个概念:广义Transformer和狭义Transformer,狭义Transformer指Attention is all you need那个结构,包含一个encoder一个decoder;广义Transformer指self-attention机制的各种应用。下面是我个人一些理解,感觉可能有不对或者不完整的

1。狭义Transformer在大数据的情况下超过RNN一点问题没有,而且是显著的超,目前在很多大数据任务都验证了,比如WMT数据集的机器翻译,BERT,以及最近的TTS。

2。狭义的Transformer在跑中等数据集(一百万)或者小数据(几万或者几千)和RNN谁好谁坏不一定,可能靠调,不是那种一个默认参数就会比RNN好,我简单的跑过中等数据集的对话(Twitter)和摘要(CNN DailyMail),感觉没好多少。小数据反而更容易过拟合,按照一些paper报的结果反而RNN好

3。广义的Transformer又是另一个故事了,我个人理解self attention机制几乎在所有任务上好好调都有用,比如分类,阅读理解(QANet),小规模s2s(Universal Transformer)。比如QAnet和Universal Transformer大家如果有兴趣可以看一看,通过自己的Recurrent机制,能给性能带来很大的增益。Multi head attention和简单的Self attention都是很有用的结构上的trick

4。我忘了是ACL的哪篇文章了,做过一个实验。RNN对于50个词之前的词顺序就不敏感了,而对于100个词之前的就完全忘了,所以RNN在长序列上肯定是有问题的。不过Transformer的位置信息只靠Position Embedding,我个人觉得可能也有提升空间


<pre style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-right: 8px;margin-left: 8px;max-width: 100%;letter-spacing: 0.544px;white-space: normal;color: rgb(0, 0, 0);font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;widows: 1;line-height: 1.75em;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;box-sizing: border-box !important;overflow-wrap: break-word !important;">—</span></strong>完<strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;font-size: 16px;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;letter-spacing: 0.5px;box-sizing: border-box !important;overflow-wrap: break-word !important;">—</span></strong></span></strong></span></strong></section><section style="max-width: 100%;letter-spacing: 0.544px;white-space: normal;font-family: -apple-system-font, system-ui, "Helvetica Neue", "PingFang SC", "Hiragino Sans GB", "Microsoft YaHei UI", "Microsoft YaHei", Arial, sans-serif;widows: 1;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section powered-by="xiumi.us" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-top: 15px;margin-bottom: 25px;max-width: 100%;opacity: 0.8;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;letter-spacing: 0.544px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section powered-by="xiumi.us" style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-top: 15px;margin-bottom: 25px;max-width: 100%;opacity: 0.8;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><section style="margin-right: 8px;margin-bottom: 15px;margin-left: 8px;padding-right: 0em;padding-left: 0em;max-width: 100%;color: rgb(127, 127, 127);font-size: 12px;font-family: sans-serif;line-height: 25.5938px;letter-spacing: 3px;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(0, 0, 0);box-sizing: border-box !important;overflow-wrap: break-word !important;"><strong style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;font-size: 16px;font-family: 微软雅黑;caret-color: red;box-sizing: border-box !important;overflow-wrap: break-word !important;">为您推荐</span></strong></span></section><section style="margin: 5px 8px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(87, 107, 149);font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">人工智能领域最具影响力的十大女科学家</span><br style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"  /></section><section style="margin: 5px 8px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(87, 107, 149);font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">MIT最新深度学习入门课,安排起来!</span></section><section style="margin: 5px 8px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(87, 107, 149);font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">有了这个神器,轻松用 Python 写个 App</span></section><section style="margin: 5px 8px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="max-width: 100%;color: rgb(87, 107, 149);font-size: 14px;box-sizing: border-box !important;overflow-wrap: break-word !important;">「最全」实至名归,NumPy 官方早有中文教程</span><br style="max-width: 100%;box-sizing: border-box !important;overflow-wrap: break-word !important;"  /></section><section style="margin: 5px 8px;padding-right: 0em;padding-left: 0em;max-width: 100%;min-height: 1em;font-family: sans-serif;letter-spacing: 0px;opacity: 0.8;line-height: normal;box-sizing: border-box !important;overflow-wrap: break-word !important;"><span style="font-size: 14px;">10个必会的 PyCharm 技巧</span></section></section></section></section></section></section></section></section></section>

有了Transformer框架后是不是RNN完全可以废弃了?

本篇文章来源于: 深度学习这件小事

本文为原创文章,版权归所有,欢迎分享本文,转载请保留出处!

知行编程网
知行编程网 关注:1    粉丝:1
这个人很懒,什么都没写

发表评论

表情 格式 链接 私密 签到
扫一扫二维码分享