1、非专业人士翻译，如有错误请谅解。Googles AI Reasons Its Way around the London Underground谷歌人工智能推导出环绕伦敦地铁系统的路线DeepMinds latest technique uses external memory to solve tasks that require logic and reasoninga step toward more humanlike AI深度思维最新技术使用了外部存储来解决需要逻辑思维和推理能力的任务By Elizabeth Gibney, Nature magazine on October 14
2、, 2016伊丽莎白.吉布尼 2016 年 10 月 14 日发表于自然杂志Artificial-intelligence (AI) systems known as neural networks can recognize images, translate languages and even master the ancient game of Go. But their limited ability to represent complex relationships between data or variables has prevented them from conquer
3、ing tasks that require logic and reasoning.人工智能(AI)系统被认为是神经网络，可以识别图片，翻译，甚至精通古老的游戏。但他们描绘数据或变量之间的复杂关系的能力有限，这妨碍了他们克服需要逻辑思维和推理能力的任务。In a paper published in Nature on October 12, the Google-owned company DeepMind in London reveals that it has taken a step towards overcoming this hurdle by creating a neur
4、al network with an external memory. The combination allows the neural network not only to learn, but to use memory to store and recall facts to make inferences like a conventional algorithm. This in turn enables it to tackle problems such as navigating the London Underground without any prior knowle
5、dge and solving logic puzzles. Though solving these problems would not be impressive for an algorithm programmed to do so, the hybrid system manages to accomplish this without any predefined rules.在 10 月 12 日自然杂志中发表的一篇论文中，谷歌在伦敦的子公司深度思维展示了他们通过结合外部存储创造了一个神经网络，来进一步克服这些障碍。这种和外部存储的结合不仅允许神经网络学习，还可以通过存储器来存
6、储和回忆事件，并以此来像正常情况那样做推断。这反过来能够让它解决难题，比如在没有任何经验的情况下操控伦敦地铁，比如解决逻辑谜题。尽管对于一个算法程序来说做到这点并不会令人印象深刻，但这个混合系统在没有任何先决条件的情况下做到了这点。Although the approach is not entirely newDeepMind itself reported attempting a similar feat in a preprint in 2014“the progress made in this paper is remarkable”, says Yoshua Bengio, a
7、computer scientist at the University of Montreal in Canada.虽然这个方法不是一个全新的技术深度思维自己就在 2014年报告过他们尝试了一种相似的技术但“在论文中的这个进步是非凡的”，加拿大蒙特利尔的计算机学家本吉奥.本希奥赞叹道。MEMORY MAGIC记 忆 魔 法A neural network learns by strengthening connections between virtual neuron-like units. Without a memory, such a network might need to se
8、e a specific London Undeground map thousands of times to learn the best way to navigate the tube.神经网络通过加强虚拟神经元之间的联系来学习。如果没有存储器，这样一个网络可能需要看一副特定的伦敦地铁地图数千次来学习最佳路线。DeepMinds new systemwhich they call a differentiable neural computercan make sense of a map it has never seen before. It first trains its ne
9、ural network on randomly generated map-like structures (which could represent stations connected by lines, or other relationships), in the process learning how to store descriptions of these relationships in its external memory as well as answer questions about them. Confronted with a new map, the D
10、eepMind system can write these new relationshipsconnections between Underground stations, in one example from the paperto memory, and recall it to plan a route.深度思维的新系统他们称它为微分神经计算机可以理解它从未见过的地图。第一次训练神经网络是在随机生成的类似结构的地图上(被铁路线链接的车站，或者其他关系)，在这个过程中学习如何将这些关系的描述存储在它的外部存储器并且回答问题。面对一个新的地图，深度思维的系统可以把这些新关系按照一个图
11、纸上例子来连接各地铁站之间的关系写到存储器，并能够回忆这些关系然后计划路线。DeepMinds AI system used the same technique to tackle puzzles that require reasoning. After training on 20 different types of question-and-answer problems, it learnt to make accurate deductions. For example, the system deduced correctly that a ball is in a playg
12、round, having been informed that “John picked up the football” and “John is in the playground”. It got such problems right more than 96% of the time. The system performed better than recurrent neural networks, which also have a memory, but one that is in the fabric of the network itself, and so is l
13、ess flexible than an external memory.深度思维的人工智能系统使用同样的方法来处理需要推理能力的智力游戏。在通过 20 种不同类型的问答训练之后，它学会了做出准确的推论。例如，系统通过被告之“约翰抓着足球”和“约翰在操场上”准确的推断出一个球在操场上。答对问题的概率超过了 96%。这个系统的效率比拥有一个内部存储器的周期神经网络更高，也更灵活。Although the DeepMind technique has proven itself on only artificial problems, it could be applied to real-wor
14、ld tasks that involve making inferences from huge amounts of data. This could solve questions whose answers are not explicitly stated in the data set, says Alex Graves, a computer scientist at DeepMind and a co-author on the paper. For example, to determine whether two people lived in the same count
15、ry at the same time, the system might collate facts from their respective Wikipedia pages.虽然深度思维的技术已被证明只针对人工问题，但它能够被应用到需要通过海量数据来进行推断的真实世界的工作。这能够解决那些在数据中没有明确答案的问题。来自深度思维的计算机科学家，研究报告的合著者，亚历克斯格雷夫斯介绍说。例如，对于判断两人是否在同一时间住在同一个国家，系统可能会核对他们各自在维基百科上的事项。Although the puzzles tackled by DeepMinds AI are simple, B
16、engio sees the paper as a signal that neural networks are advancing beyond mere pattern recognition to human-like tasks such as reasoning. “This extension is very important if we want to approach human-level AI.”虽然对于深度思维的人工智能来说，智力游戏很简单，但本希奥认为该论文是一个信号，它表明神经网络正在跨越单纯的模式识别，成长到能够做人类才能做的任务，例如推理。“如果我们想实现像人一样的人工智能，这次突破是非常重要的。”This article is reproduced with permission and was first published on October 13, 2016.这篇文章允许被转载，首次发表是 2016 年 10 月 13 日。
