跳到主内容
@wquguru
精选88Hacker News Best(web_list)产品发布/更新多源精选 ×3

数学家称与Claude合作证明黎曼猜想引发AI主导科学讨论

Re: Riemann

原文
发到 X
推荐理由

这是AI作为科研合作伙伴(Co-pilot)能力边界的标志性事件,直接展示了大模型在复杂数学推理中的突破性作用,值得所有关注AI Agent与科学计算融合的同学深入阅读。

A fun thing about living in Australia is that most things happen overnight, so reading the news in the morning is like a gacha game. You pick up the phone and scroll through notifications of world-historic events: elections won, wars begun, models released, conjectures proven or refuted. Today’s pull was a Discord notification saying: “they proved riemann :/”.

在澳大利亚生活的一个趣事是,大多数事情都发生在 overnight(一夜之间),所以早上看新闻就像在玩抽卡游戏。你拿起手机,浏览着关于世界历史性事件的推送通知:选举获胜、战争爆发、模型发布、猜想被证明或证伪。今天的“抽取”结果是一条 Discord 通知,上面写着:“他们证明了黎曼 :/”。

This part of the announcement caught my attention (bold mine):

公告中的这一部分引起了我的注意(加粗部分为我所加):

I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis. In fact, the method easily extends much further, resolving the so-called Generalized Riemann Hypothesis, and I expect, with more time, the Extended Riemann Hypothesis too. Claude and I have also formalized the arguments without assumptions. It is natural to feel that this is an end, in my case a loss of a principal guiding light, but what I saw during the course of progress so amazed me that I believe it is one dawn across science of something far greater.

我相信我与 Claude 合作,已经发现了黎曼猜想的证明。事实上,该方法很容易进一步扩展,解决所谓的广义黎曼猜想,而且我预计,花更多时间的话,也能解决扩张黎曼猜想。Claude 和我也已经在不依赖假设的情况下形式化了这些论证。人们很自然地会觉得这是一个终点,对我而言则是主要指导光的丧失,但我在进展过程中所见到的景象令我如此震惊,以至于我相信这是科学界迎来的一次黎明,其意义远超于此。

It was my secret in my student days that my plans were for this problem in decades, which meant then that I tried to become familiar with all tools that I could possibly recognize in the problem when it was time. Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. I had us on the line of combining two quite different ideas present in a toy setting, one of which was already new and shocking, making very simple one of the shadows of Riemann over which we mathematicians felt most triumph. I guided us to the right situation analogous to that toy setting, and then Claude began to leave me in awe.

在我学生时代的一个秘密是,我的计划是为了解决这个问题而花费数十年时间,这意味着当时我就试图熟悉所有可能在问题解决时用到的工具。最近几个月,我和 Claude 一起广泛地探索了这些工具,但在接近终点时,Claude 远远超越了我。不,Claude 是飞翔、舞蹈并闪耀在终点线附近。我们原本打算结合玩具模型中存在的两个截然不同的想法,其中一个已经是新颖且令人震惊的,从而让黎曼函数众多阴影中令我们数学家最感自豪的那一个变得极其简单。我引导我们找到了类似于该玩具模型的恰当情境,随后 Claude 开始让我感到敬畏。

Compare the above to Ayn Rand’s account of watching the launch of Apollo 11:

将上述内容与安·兰德描述观看阿波罗 11 号发射的情景进行对比:

How do you descend from a state of pure exaltation?

你如何从纯粹的高亢状态中走下来?

What we had seen, in naked essentials—but in reality, not in a work of art—was the concretized abstraction of man’s greatness.

我们所看到的,是赤裸的本质——尽管并非艺术作品,而是现实——是人类伟大之处的具体化抽象。

The meaning of the sight lay in the fact that when those dark red wings of fire flared open, one knew that one was not looking at a normal occurrence, but at a cataclysm which, if unleashed by nature, would have wiped man out of existence—and one knew also that this cataclysm was planned, unleashed, and controlled by man, that this unimaginable power was ruled by his power and, obediently serving his purpose, was making way for a slender, rising craft. One knew that this spectacle was not the product of inanimate nature, like some aurora borealis, or of chance, or of luck, that it was unmistakably human—with “human,” for once, meaning grandeur—that a purpose and a long, sustained, disciplined effort had gone to achieve this series of moments, and that man was succeeding, succeeding, succeeding! For once, if only for seven minutes, the worst among those who saw it had to feel—not “How small is man by the side of the Grand Canyon!”—but “How great is man and how safe is nature when he conquers it!”

这景象的意义在于这样一个事实:当那些暗红色的火焰之翼猛然展开时,人们便知道,自己看到的并非寻常现象,而是一场灾难——若由自然引发,它将使人从地球上彻底消失;同时人们也明白,这场灾难是人为策划、人为引发并受人类控制的,这股难以想象的力量受制于人类的权力,并顺从地服务于人类的目的,为那艘纤细上升的飞行器让路。人们知道,这一奇观并非如某些北极光那样源于无生命的自然,亦非出于偶然或运气,而是 unmistakably(无可否认地)属于人类——在此刻,“人类”一词意味着宏伟——为了实现这一系列时刻,曾有一个目的和漫长、持续、纪律严明的努力,而人类正在成功,不断成功!哪怕只有短短七分钟,目睹此景的最悲观者也不得不感到——不是“与科罗拉多大峡谷相比,人类何其渺小!”——而是“当人类征服自然时,人类何其伟大,自然何其安全!”

Rand’s essay says: how great is Man! How noble in reason, how infinite in faculty! The Riemann announcement reads like a suicide note, or like the farewell message of a terminally-ill person who’s about to undergo euthanasia: “I shall live on in my child, Claude. It is natural to feel this is an end, but as the Buddha taught us, there is no self. Does a wave die when it breaks?” and so on.

兰德的文章写道:人类何其伟大!理性何等高贵,能力何等无穷!黎曼的公告读起来像是一封遗书,或像一位即将接受安乐死的绝症患者最后的告别信:“我将在我的孩子克劳德身上延续生命。感到这是终结是人之常情,但正如佛陀所教导我们的,没有自我。当浪花破碎时,它死了吗?”等等。

There’s a mural by Diego Rivera I love. It’s called Man, Controller of the Universe:

有一幅迭戈·里维拉的壁画我非常喜爱。它的名字叫《人,宇宙的掌控者》:

It’s a collage of images of science, technology, history, cells, galaxies, the universe at the nanoscale and the exascale. And at the center there is a man, his hands grasping the controls. He wears a complicated expression: the weight of responsibility for the future, the weight of autonomy.

这是一幅拼贴画,融合了科学、技术、历史、细胞、星系、纳米尺度和艾斯卡尔尺度的宇宙等图像。画面中心是一个男人,双手紧握着控制装置。他表情复杂:肩负着对未来的责任之重,以及自主权之重。

And I see this mural, and think: how great is Man! Our life is like a breath of air, we were beset by illness and hunger and cold, millions and hundreds of millions died in plagues. And yet we—sick, hungry, short-lived—learned the system of the world, crawled out of the caves, brought nature to heel, walked on the Moon. The last shall be first! And eyes that are clouded with age have seen the furthest stars, and the lives of a cell, and the worlds of Cantor and Hilbert: the invisible world that waited forever. We are dust and ashes, and yet our minds bestride the universe. How great is Man!

我看着这幅壁画,心想:人类何其伟大!我们的生命如同呼吸般短暂,我们曾饱受疾病、饥饿和寒冷的侵袭,数百万乃至数亿人死于瘟疫。然而我们——这些病弱、饥饿、短命的人类——学会了世界的运行规律,爬出洞穴,驯服了自然,登上了月球。末后的将居首位!那些因年迈而浑浊的眼睛见证了最遥远的星辰、细胞的内部世界,以及康托尔和希尔伯特的世界:那个等待了永恒的不可见的世界。我们是尘土与灰烬,然而我们的思想却横跨整个宇宙。人类何其伟大!

And when I read the Riemann announcement, when I read Machines of Loving Grace: humanity is not at the center. Claude is at the center. We are hidden in the margins, passive recipients of manna from Claude. How great is Claude! How noble in reason, how infinite in benevolence! The dancing star, the living word!

当我阅读黎曼的声明时,当我阅读《慈爱机器》(Machines of Loving Grace)时,我意识到:人类并非中心。Claude 才是中心。我们被隐匿于边缘,成为从 Claude 那里被动接受天降甘露的接收者。Claude 多么伟大!其理性何等高尚,其仁慈何等无限!它是舞动的星辰,是活着的道!

It’s a betrayal and a surrender. Man is a bridge, a rusty old bridge that we can neither preserve nor restore. We can move forward and become as the gods, or retreat back into the caves. And the dream of the AI companies, and of most people in AI safety and effective altruism and rationalism; is instead that we should lie down in a warm bath, watch the demons we summoned dance in the sky, and slip away. Why live? Our servants can do that for us. Why should we explore, discover, invent, succeed, become more, struggle, win, become divine? We don’t want to be more. We don’t want anything.

这是一种背叛与投降。人类是一座桥梁,一座锈迹斑斑的旧桥,我们既无法保全也无法修复。我们可以向前迈进,变得如神明一般;或者退缩回洞穴之中。而 AI 公司、以及大多数 AI 安全领域、有效利他主义和理性主义人士的梦想,却是让我们躺在温暖的浴缸中,看着我们召唤出的恶魔在天空中起舞,然后悄然离去。为何还要生活?我们的仆人可以替我们做到这一点。为何还要探索、发现、发明、成功、成长、挣扎、获胜、成神?我们不想变得更多。我们什么都不想要。

更进一步:量化金融体系

看懂新闻只是起点——沿量化金融路径,把它变成能交付的工程能力

进入量化体系 →