区块链，一种诞生于比特币的底层技术，一举成为2018年的新“风口”，其话题热度及所受关注度甚至赶超人工智能。而随着两大技术的发展，越来越多的人开始将两者相提并论，探讨人工智能与区块链融合发展的可能性。那么，当人工智能遇上区块链，这个世界将发生哪些改变？
, a low-level technology that was born in Bitcoin and became a new “windlight” in 2018, with the topic of heat and concern exceeding . As two technologies evolve, more and more people are beginning to bring the two together and explore the possibilities for the integration of artificial intelligence into the chain.
人工智能和区块链的共同点
common elements of artificial intelligence and block chains
区块链关注的是保持准确的记录、认证和执行，而人工智能则助力于决策、评估和理解某些模式和数据集，最终产生自主交互。人工智能和区块链共同拥有几个特点，可以确保在不久的将来能够实现无缝互动。下面列出了三个主要特点。
Block chains are concerned with keeping accurate records, authentication and enforcement, while artificial intelligence supports decision-making, assessment and understanding of certain models and data sets, leading to autonomous interactions. Artificial intelligence and block chains share several features that can ensure seamless interaction in the near future.
1. 人工智能和区块链需要数据共享
1. Artificial intelligence and block chains require data sharing
分布式数据库强调了在特定网络上的多个客户端之间共享数据的重要性。同样，人工智能依靠大数据，特别是数据共享。可供分析的开放数据越多，机器的预测和评估则会更加正确，生成的算法也更加可靠。
Distributed databases emphasize the importance of sharing data between multiple clients on a given network. Similarly, artificial intelligence relies on large data, especially data sharing.
2. 安全
Security
处理区块链网络上进行高价值交易时，这对网络的安全性有很大的要求。这可通过现有协议实施。对于人工智能来说，机器的自主性也需要很高的安全性，以降低发生灾难性事件的可能性。
When dealing with high-value transactions on the block chain network, this is very much a requirement for the safety of the network. This can be done through existing protocols. For artificial intelligence, the autonomy of the machine also requires a high degree of security in order to reduce the likelihood of catastrophic events.
3. 信任是必要条件
3. Trust as a sine qua non
对于任何广泛接受的技术的进步，没有比缺乏信任具有更大的威胁，也不排除人工智能和区块链。为了使机器间的通信更加方便，则需要有一个预期的信任级别。想要在区块链网络上执行某些交易，信任则是一个必要条件。
There is no greater threat to the advancement of any widely accepted technology than a lack of trust, nor does it exclude artificial intelligence or block chains. To make communication between machines easier, an expected level of trust is required. Trust is a necessary condition for certain transactions to be carried out on the block chain network.
区块链如何改变人工智能
上面介绍了几个人工智能和区块链具有的相同特点，现在继续了解区块链如何改变人工智能。
The same characteristics as those of several artificial intelligence and block chains are described above, and there continues to be an understanding of how the block chain can change artificial intelligence.
1．开放的数据市场
1. Open data market
如前所述，人工智能技术的进步取决于各种来源数据的可用性。尽管像谷歌，Facebook，亚马逊等这样的公司可以访问大量的人工智能数据源，这些数据对于大部分人工智能应用也都非常有用，但在数据市场上并不能对这些数据进行直接访问。
As noted earlier, advances in AI technology depend on the availability of data from various sources. While companies like Google, Facebook, Amazon, etc. have access to a large number of AI data sources, these data are also useful for most AI applications, but they cannot be accessed directly in the data market.
区块链旨在通过引入点对点连接这一概念来解决这个问题。由于它是一个开放的分布式注册表，因此网络上的每个人都可以访问数据。现有的数据寡头垄断即将结束，一个新的开放和自由数据的时代即将来临。
The block chain is designed to solve this problem by introducing the concept of point-to-point connection. Because it is an open distributed register, everyone on the network has access to data. The existing data oligopoly is coming to an end, and a new era of open and free data is approaching.
2.大规模的数据管理机制
2. Large-scale data management mechanisms
即使数据已经可以对所有人都开放，对数据的管理也是另一障碍。目前可用的数据量约为1.3泽字节(Zettabytes)。人工智能的一个子领域称为通用人工智能(Artificial General Intelligence)，它可以建立一个反馈控制系统的模型，有助于自主代理人(autonomous agents)与物理环境更好地进行交互。
Even if the data are already accessible to all, the management of the data is another obstacle. The amount of data currently available is about 1.3 zettabytes. One sub-area of artificial intelligence, known as Artificial General Intelligence, allows for a model of a feedback control system that allows for better interaction between autonomous agents (automoous agents) and the physical environment.
与传统中央存储中心相比，具有大量数据存储的分布式系统享有多种优势。在发生危机和自然灾害时，数据没有存储在单个位置，因此数据可以受到保护。此外，网络被黑客攻击的行为没有了，这使得数据集不易受到部分损坏的影响。
Data can be protected when they are not stored in a single location during a crisis or natural disaster. Moreover, network attacks by hackers are absent, making data sets less vulnerable to partial damage.
3.更可靠的人工智能建模和预测
3. More reliable modelling and forecasting of artificial intelligence
计算机系统的一个基本原则是GIGO：垃圾进垃圾出。人工智能领域严重依赖于大量的数据流，一些个人或公司故意篡改提供的数据以期待改变结果，垃圾数据也可能是由传感器和其他数据源的意外故障引起的。
One of the basic principles of computer systems is GIGO: Garbage in and out of waste. The area of artificial intelligence is heavily dependent on a large flow of data, some individuals or companies deliberately alter the data provided in anticipation of a change in results, and garbage data can also be caused by accidental malfunctions in sensors and other data sources.
通过创建已验证数据库的各个部分，可以成功构建模型并仅在已验证的数据集上实施。这将检测数据供应链中的任何故障或意外情况。由于数据流部分可用，因此它还有助于降低故障排除和查找异常数据集的压力。最后，区块链与不变性同义，这意味着数据是可追踪和可审查的。
By creating various parts of a validated database, models can be successfully constructed and implemented only on validated data sets. This will detect any malfunctions or contingencies in the data supply chain. It will also help to reduce the pressure on troubleshooting and searching for unusual data sets because the data stream is available. Finally, block chains are synonymous with non-variability, which means that the data are traceable and revegetable.
4.对数据和模型使用的控制
4. Controls over the use of data and models
这是整合区块链和人工智能的一个非常重要的方面。例如，当你登录Facebook和Twitter时，你将会放弃将资源上传到其平台上的权利。当歌手签署唱片协议时也会发生同样的事情。相同的概念也可以应用于人工智能数据和模型。
This is a very important aspect of integrating block chains and artificial intelligence. For example, when you log on Facebook and Twitter, you will give up the right to upload resources onto their platforms. The same thing happens when singers sign record agreements. The same concept can be applied to artificial intelligence data and models.
为建模创建数据时，你可以指定有限制或许可的许可证。区块链使得这一过程变得相对容易了一些。
When creating data for modelling, you can specify a restricted or licensed licence. Block chains make the process relatively easy.
为了解释在区块链网络中查看或使用数据的权限被视为一项资产。与硬币可以在加密货币平台上传递的方式一样，这些访问网络信息的权限也可以进行传递。
The permission to view or use data in a block chain network is considered an asset. As with coins that can be transmitted on an encrypted monetary platform, these access rights to network information can also be passed on.
技术共赢：人工智能与区块链互解难题
technology wins: an artificial intelligence and block chain solver puzzle
业界人士认为，区块链技术与人工智能达成合作的潜力巨大，二者可以互相解决技术难题，在新的生态建构中，数据存储、共享机制、平台问题、安全性问题等，都可以利用彼此的技术实现攻坚克难。例如在数据领域，AI可以与区块链技术结合，一方面是从应用层面入手，两者各司其职，AI负责自动化的业务处理和智能化的决策，区块链负责在数据层提供可信数据；另一方面是数据层，两者可以互相渗透。区块链中的智能合约实际上也是一段实现某种算法的代码，既然是算法，那么AI就能够植入其中，使区块链智能合约更加智能。同时，将AI引擎训练模型结果和运行模型存放在区块链上，就能够确保模型不被篡改，降低了AI应用遭受攻击的风险。
In the field of data, for example, AI can combine the technology of block chains with the application level on the one hand, AI is responsible for automated business processing and intelligent decision-making, and AI is responsible for providing credible data on the data layer on the other hand, and the data layer on the other. Smart contracts in block chains are actually a code for achieving some algorithms, and since algorithms, AI can implant them to make block chain smart contracts more intelligent. At the same time, storing the AI engine training model results and operating models on sector chains ensures that models are not tampered with and reduces the risk of AI applications being attacked.
如果说人工智能是一种生产力，它能提高生产的效率，使得我们更快、更有效地获得更多的财富。那么区块链就是一种生产关系，它能够改变我们一些分配。人工智能和区块链能够基于双方各自的优势实现互补。事实上，目前业界已经有公司尝试将两者同时应用。
If artificial intelligence is a productivity that increases productivity and enables us to gain more wealth faster and more efficiently, then block chains are a productive relationship that can change some of our distributions. Artificial intelligence and block chains can complement each other’s respective strengths.
注册有任何问题请添加 微信：MVIP619 拉你进入群
打开微信扫一扫
添加客服
进入交流群
发表评论