2022年经济学人 人工智能在策略游戏中胜过人类(2)

The Meta team's crucial contribution was therefore to augment¹ reinforcement learning with natural-language processing.

因此，该团队做出的关键贡献是利用自然语言处理来增强强化学习。

Large language models, trained on vast amounts of data to predict deleted words, have an uncanny ability to mimic² the patterns of real language and say things that humans might.

以海量数据为基础进行训练，预测删除词的大语言模型具有一种不可思议的能力，可以模仿人类语言的模式，说出人类可能说的话。

For Cicero, the team started with a pre-trained model with a baseline understanding of language, and fine-tuned this on dialogues from more than 40,000 past games, to teach it Diplomacy³-specific patterns of speech.

就Cicero来说，团队从对语言有基本理解能力的预训练模型开始，利用过去超过4万场《强权外交》游戏比赛的对话对其进行扩充，以教授它该游戏特有的言语模式。

To play the game, Cicero looks at the board, remembers past moves and makes an educated guess as to what everyone else will want to do next.

在这个游戏中，Cicero需要看着棋盘，记住自己此前的行动，并对其他人的下一步行动做出合理推测。

Then it tries to work out what makes sense for its own move, by choosing different goals, simulating what might happen, and also simulating how all the other players will react to that.

然后，它尝试选择不同的目标，模拟可能发生的情况，以及所有其他参与者对此做出的反应，来判断行动是否合理。

Once it has come up with a move, it must work out what words to say to the others.

一旦决定了如何行动，它要认真思考如何与其他人对话。

To that end, the language model spits out possible messages, throws away the bad ideas and anything that is actual gobbledygook, and chooses the ones, appropriate to the recipients⁴ concerned, that its experience and algorithms suggest will most persuasively⁵ further its agenda.

为此，语言模型会给出可能的信息，丢弃糟糕的想法和官话，选择最能赢得对手青睐的想法，这些想法是模型根据其经验和算法得出的最有说服力，且最有可能推进其议程的。

Cicero, then, can negotiate, convince, cooperate and compete.

然后，Cicero可以与对手进行谈判、游说、合作和竞争。

Seasoned Diplomacy players will, though, want to know something else: has it learned how to stab?

然而，经验丰富的《强权外交》玩家还想知道一件事：它学会如何背刺了吗？

Stabbing - saying one thing and doing another (especially, attacking a current ally) is seen by many as Diplomacy's defining feature.

背刺——说一套做一套（尤其是攻击当前的盟友）被许多人视为《强权外交》游戏中的定义性特征。

But, though Cicero did, “strategically withhold⁶ information from players in gameplay”, it did not actually stab any of its opponents.

尽管Cicero确实“在战略上对其它玩家隐瞒了信息”，但它实际上没有暗中伤害任何对手。

Perhaps it was this final lack of Machiavellian⁷ ruthlessness which explains why it was only in the top 10%, and not victor ludorum.

也许正是由于缺失这种马基雅维利主义的冷酷无情，解释了为什么Cicero只进入了排名前10%，而没有最终成为“总冠军”。

1 augment   

vt.(使)增大，增加，增长，扩张

参考例句：

• They hit upon another idea to augment their income.他们又想出一个增加收入的办法。
• The government's first concern was to augment the army and auxiliary forces.政府首先关心的是增强军队和辅助的力量。

2 mimic   

v.模仿，戏弄；n.模仿他人言行的人

参考例句：

• A parrot can mimic a person's voice.鹦鹉能学人的声音。
• He used to mimic speech peculiarities of another.他过去总是模仿别人讲话的特点。

3 diplomacy   

n.外交；外交手腕，交际手腕

参考例句：

• The talks have now gone into a stage of quiet diplomacy.会谈现在已经进入了“温和外交”阶段。
• This was done through the skill in diplomacy. 这是通过外交手腕才做到的。

4 recipients   

adj.接受的；受领的；容纳的；愿意接受的n.收件人；接受者；受领者；接受器

参考例句：

• The recipients of the prizes had their names printed in the paper. 获奖者的姓名登在报上。 来自《简明英汉词典》
• The recipients of prizes had their names printed in the paper. 获奖者名单登在报上。 来自《现代英汉综合大词典》

5 persuasively   

adv.口才好地；令人信服地

参考例句：

• Students find that all historians argue reasonably and persuasively. 学生们发现所有的历史学家都争论得有条有理，并且很有说服力。 来自辞典例句
• He spoke a very persuasively but I smelled a rat and refused his offer. 他说得头头是道，但我觉得有些可疑，于是拒绝了他的建议。 来自辞典例句

6 withhold   

v.拒绝，不给；使停止，阻挡

参考例句：

• It was unscrupulous of their lawyer to withhold evidence.他们的律师隐瞒证据是不道德的。
• I couldn't withhold giving some loose to my indignation.我忍不住要发泄一点我的愤怒。

7 machiavellian   

adj.权谋的，狡诈的

参考例句：

• A Machiavellian plot was suspected.人们怀疑背后有不可告人的阴谋。
• In this layer,Obama implied American policies that are cautious and Machiavellian.在这个层面，奥巴马含蓄地表达了美国的谨慎、权谋的（新）政策。