Name: "我在薄冰上滑行了 40 年 ("I've Been Skating On Thin Ice For 40 Years")
Uploaded: 2025-06-17T16:53:00.000Z
Duration: 12 min 18 s

程度副詞

In celebration of World Statistics Day we're bringing you some of the most fascinating data-driven stories. From using math to skate on thin ice to AI poker and even a James Bond villain lair. We've got it all. What would your ideal villain lair look like? Let us know in the comments. The ice is typically around five centimeters when I skate on it. That sound you hear is actually a combination of two sounds. It's the sounds from cracks striking beneath me and the vibrations within the ice plate itself. It sounds like it's something supersonic.

為慶祝世界統計日，我們將為您帶來一些最精彩的數據驅動故事。從利用數學如履薄冰到人工智能撲克，甚至是詹姆斯-邦德的大反派巢穴。我們一網打盡。你理想中的反派巢穴是什麼樣的？請在評論中告訴我們。我在冰上滑行時，冰面通常只有五釐米左右。你聽到的聲音其實是兩種聲音的組合一種是我腳下的裂縫撞擊聲 另一種是冰板本身的震動聲聽起來像是超音速的聲音。

轉承詞

Meet Morten Einer. Hi, I'm Morten Einer. He's a mathematician who enjoys riskier hobbies than most. I've been skating on thin ice for 40 years. When he's skating though he's also listening.

認識莫滕-艾納。你好，我是莫頓-艾納。他是個數學家，喜歡比大多數人更冒險的愛好40年來，我一直如履薄冰。當他滑冰的時候，他也在傾聽。

情態副詞

The thinner the ice, the higher the pitch. Just about when the ice is to break, and that would be about three centimeters, the pitch is at the higher C, the supposedly highest note of the soprano C. Morten relies on his knowledge as a mathematician to calculate the thickness of the ice, its flexibility, and its temperature. Many people tend to freak out when they see us skating on thin ice, but I couldn't be more calm because I know what I'm doing. Sometimes the ice is perfectly clear and I can see fish below, even a beaver, and also freedivers swimming under the ice. When Morten first started, there wasn't any physics or scientific information to explain how thin ice skating is possible. He studied the ice for 10 years every winter to explain the phenomenon.

冰越薄，音調越高。莫頓依靠他作為數學家的知識計算出冰的厚度、彈性和溫度。很多人看到我們在薄冰上滑行時都會嚇一跳，但我卻非常鎮定，因為我知道自己在做什麼。有時冰面非常清晰，我能看到下面的魚，甚至海狸，還能看到自由潛水員在冰下游泳。莫滕剛開始滑冰時，沒有任何物理學或科學資訊來解釋薄冰滑冰是如何實現的。為了解釋這一現象，他每年冬天都要對冰面進行長達 10 年的研究。

I observe three things with a thin ice. One is that the ice bends, and the thinner it is, the more it bends. The ice needs to be flexible to support you. Second, the first crack is a true warning signal, then I know the ice is thin, and when it gets even thinner, more cracks evolve.

我觀察到薄冰有三點。一是冰會彎曲，而且越薄越彎曲。冰需要有彈性才能支撐你。第二，第一條裂縫是一個真正的警告信號，然後我就知道冰很薄，當冰變得更薄時，裂縫就會更多。

Finally, the ice gives sonorous tone, beautiful eerie tone, which immediately tells me how thick the ice is. It's called the coincidence frequency, and it can be calculated mathematically.

最後，冰發出鏗鏘有力的音調，優美陰森的音調，立刻告訴我冰有多厚。這叫重合頻率，可以用數學方法計算出來。

Morten has skated on over 1,800 bodies of water in North America,

莫滕曾在北美 1800 多個水域滑冰、

Norway, and Central and Eastern Europe. I love skating for the excitement and challenges it brings. It's an opportunity for math and nature to come together and make it understandable for me.

挪威、中歐和東歐。我喜歡滑冰帶來的刺激和挑戰。這是一個數學與自然結合的機會，讓我能夠理解數學。

I've been skating most of my life, and I always look forward to next winter.

我滑冰滑了大半輩子，一直期待著下一個冬天的到來。

Hi, welcome to Great Big Story Questions. Please state your name and your specialist subject.

你好，歡迎來到 "大故事問題"。請說出您的姓名和專業主題。

I'm Stuart, I'm from London, and my specialist subject is brilliant. Let's start the clock.

我叫斯圖爾特，來自倫敦，我的專業課程是精彩紛呈。讓我們開始計時。

Instead of memorizing, which brand uses hands-on problem solving, a learning method proven to be six times more effective than watching lectures?

與死記硬背相比，哪個品牌採用了動手解決問題的學習方法，這種學習方法被證明比觀看講座的效果高出六倍？

Pass. The answer was brilliant. Next question. Coding with Python, predicting probability, vector graphs, and improving large language models are all lessons you can find in which interactive learning app? Pass. It's brilliant. Last question. In 2023, it was found that

通過。答案很精彩。下一個問題使用 Python 進行編碼、預測概率、向量圖和改進大型語言模型都是你可以在哪個互動學習應用程序中找到的課程？通過。很精彩。最後一個問題。2023 年，人們發現

Americans spend on average 4.5 hours on their phones per day. Which platform builds your personal and professional growth by converting some of that time into learning? I really don't know. Pass. Really? It's brilliant. That's the whole point. I give up.

美國人平均每天在手機上花費 4.5 個小時。哪個平臺能將這些時間轉化為學習時間，從而促進您的個人和職業發展？我真的不知道。通過。真的嗎？太精彩了。這就是重點。我放棄

Brilliant is where you learn by doing, with thousands of interactive lessons in math, data analysis, programming, and AI. To try everything Brilliant has to offer for free for a full 30 days, visit brilliant.org slash greatbigstory, or click on the link in the description. You'll also get 20% off an annual premium subscription. Learn by doing with Brilliant.

Brilliant 讓您在實踐中學習，提供數千節數學、數據分析、編程和人工智能方面的互動課程。要在 30 天內免費試用 Brilliant 提供的所有課程，請訪問 brilliant.org slash greatbigstory，或點擊描述中的鏈接。訂閱年度高級版還可享受八折優惠。與 Brilliant 一起在實踐中學習。

您能勝過這位人工智能撲克玩家嗎？

When I look up at the stars, it's very hard not to think about big questions. But you know, it can turn I think anyone into a philosopher. And you're looking back in time, you're looking at something that's such an immense distance, your brain can't even compute it.

當我仰望星空時，很難不去思考一些重大的問題。但你知道，它可以把任何人變成哲學家你在回望過去 你在看距離如此遙遠的東西 你的大腦甚至無法計算它

I'm Liv Bury, and my educational background is in astrophysics. And I've been absolutely fascinated with space ever since I was a little girl. I grew up in Kent. And at university,

我是麗芙-布里，我的教育背景是天體物理學。我從小就對太空非常著迷。我在肯特郡長大。上大學時

I studied physics with astrophysics, using the language of mathematics to figure out mechanisms and processes behind the nature of reality itself. But now I'm a poker player, primarily.

我學的是物理學和天體物理學，用數學語言來探究現實本質背後的機制和過程。但現在我主要是個撲克玩家。

The game of No Limit Hold'em is a game of incomplete information. There is more unique situations in the game of No Limit Hold'em than there are atoms in the universe. It's crazy complicated. It's just numbers on a mind boggling scale. And that's what I love about the game. It's about being as methodical as you can, but also adding this sort of intuitive flair of how to read humans as well. Poker, and in particular, heads up No Limit Texas Hold'em poker has become the leading benchmark in the AI community for solving imperfect information games. I am Dr.

無限下注德州撲克是一種資訊不完全的遊戲。無限下注德州撲克遊戲中的獨特情況比宇宙中的原子還要多。複雜得令人發瘋。這只是令人匪夷所思的數字。而這正是我喜歡這個遊戲的原因。它讓你儘可能有條不紊，但同時也加入瞭如何解讀人類的直覺天賦。撲克，尤其是德州撲克的 "單挑無上限 "已成為人工智能界解決不完全資訊遊戲的領先基準。我是

Tuomas Sandholm. I'm a professor in the computer science department at Carnegie Mellon University.

圖馬斯-桑德霍爾姆我是卡內基梅隆大學計算機科學系的教授。

We developed the strategy for Libratus. Libratus is the first poker AI run on Bridges, which is the newest supercomputer at the Pittsburgh Supercomputing Center. Bridges is the largest system in the world that converges high-performance computing with artificial intelligence and big data. I'm Nick Nystrom. I'm interim director at the Pittsburgh Supercomputing

我們為 Libratus 開發了策略。Libratus 是首次在匹茲堡超級計算中心最新超級計算機 Bridges 上運行的撲克人工智能。Bridges 是世界上最大的高性能計算與人工智能和大數據融合的系統。我是尼克-尼斯特羅姆。我是匹茲堡超級計算中心的臨時主任。

Center. Pairing Libratus with the HPE Supercomputer Bridges, we really made history.

中心。將 Libratus 與 HPE 超級計算機橋配對使用，我們確實創造了歷史。

I remember hearing about a team of some of the very best No Limit heads up players were going to be playing against Libratus. And we all watched it fairly closely to see how a human mind is matched up against a supercomputer. A key aspect is that there are both multiple players and missing information. The algorithms automatically created the poker strategy for Libratus. They beat the humans in real time at the game they're best at. That was a groundbreaking moment here.

我記得曾聽說過一支由最優秀的無限下注玩家組成的隊伍將與 Libratus 對戰。我們都相當認真地觀看了比賽，想看看人類大腦是如何與超級計算機對抗的。關鍵的一點是，既有多名玩家，又有資訊缺失。算法自動為 Libratus 制定了撲克策略。它們在人類最擅長的遊戲中實時擊敗了人類。這是一個開創性的時刻。

To me, it was a demonstration of the power of the human mind because it takes a supercomputer to start beating some of the best minds in poker. And now the question is, where do we take it from there? What else do we apply it to? Like you can apply science to poker, you can apply science to fixing the world's problems. The algorithms in Libratus can be applied to situations where we're missing information, precision medicine, global hunger, climate, environmental damage. The possibilities really are limited only by our imagination. What I'm most personally excited about is the application of supercomputing in AI to solve some of these really big questions that we have about the nature of the universe itself. In some ways we are tiny and insignificant, but at the same time we also have incredible potential with our minds and the things that we can do with them.

對我來說，這展示了人類思維的力量，因為超級計算機才能開始擊敗撲克界的一些頂尖高手。現在的問題是，我們該把它帶到哪裡去？我們還能把它應用到什麼地方？就像你可以將科學應用於撲克一樣，你也可以將科學應用於解決世界問題。Libratus 中的算法可以應用於我們缺少資訊的情況、精準醫療、全球飢餓、氣候、環境破壞。我們的想象力真的很有限。最讓我個人感到興奮的是，超級計算在人工智能中的應用，可以解決我們對宇宙本身性質的一些重大問題。在某些方面，我們是渺小和微不足道的，但與此同時，我們的大腦和我們能做的事情也具有驚人的潛

人工智能也在提高人類的工作效率。

Ground reaction force data, joint kinematics, underfoot pressure, thermal cameras, high-speed video. The bridge between data and performance is really that with the right information we can make better decisions.

地面反作用力數據、關節運動學、腳下壓力、熱像儀、高速視頻。數據與性能之間的橋樑實際上是，有了正確的資訊，我們就能做出更好的決策。

Hi, I'm Keith Stern and I use data to help athletes break records. At the top level runners there's only a few seconds between the person that finishes first and the person that doesn't podium finish at all, so any small change can really make a big impact down the road.

大家好，我是基思-斯特恩，我利用數據幫助運動員打破紀錄。在頂級選手中，獲得第一名的選手和沒有登上領獎臺的選手之間只有幾秒鐘的差距，是以任何微小的改變都會對未來產生巨大的影響。

My job here is to work with the rest of the Reebok Innovation team to help bring data into the design process.

我在這裡的工作是與銳步創新團隊的其他成員合作，幫助將數據引入設計流程。

Hi, my name is Dan Hobson. I'm the VP of Innovation. There's a lot of information to take into consideration on campus. Here we built a CrossFit box and lived with the athletes monitoring how they were working out, what they were eating, how they were recovering, and making that understandable for the design team.

大家好，我是丹-霍布森（Dan Hobson）。我是創新部副總裁。在校園裡有很多資訊需要考慮。在這裡，我們建造了一個 CrossFit 健身房，與運動員一起生活，監督他們如何鍛鍊、吃什麼、如何恢復，並讓設計團隊瞭解這些資訊。

Keith Stern is testing the product. He's gathering data. The amount of data he can compile with his team is vast. A lot of our testing on athletes starts here in the Human Performance Engineering

基思-斯特恩正在測試產品。他在收集數據。他和他的團隊可以彙編大量數據。我們對運動員進行的大量測試都是從人體機能工程部開始的。

Lab. The force plate treadmill is actually a split belt treadmill so that we can record forces from the left and right foot.

實驗室。受力板跑步機實際上是一個分帶跑步機，這樣我們就可以記錄來自左右腳的力。

Our mechanical impacting machine simulates what we see in ground reaction force data and then look at the performance of material over time.

我們的機械衝擊機可以模擬地面反作用力數據，然後觀察材料隨時間變化的性能。

Through high-speed video we can track how much your foot is moving in certain areas.

通過高速視頻，我們可以追蹤到您的腳在某些部位的移動程度。

A lot of times we can't see with the naked eye what's going on in those really fast movements, so slowing it down, that transfer of knowledge is way easier than graphs and charts and numbers. It's really easy to understand how the shoe works with that person can make a big difference in the running form. The future of data collection and athletic performance is something that's really important. There's a growing field of smart sensors. Sensors could be woven into this technology that we're developing with FlexWeave. FlexWeave is really a method of weaving materials together and allows us to combine structure or durability into a single package that's also still lightweight and breathable. The same types of things that we record in the lab, we could then record on foot in the field. So if you have those smart fibers in the FlexWeave, if you want to track heart rate or distance or put GPS in it, you know, into those fibers, you can weave that into the fabric. The more data we have, the better we can make decisions on what types of material advances are actually making an improvement. Providing that performance benefit, even if it's just comfort to get people up and active, that's what's really important to me.

很多時候，我們無法用肉眼看到那些快速運動中發生了什麼，所以放慢速度，知識的傳遞要比圖表和數字容易得多。我們很容易就能理解鞋子如何與人配合，從而使跑步姿勢大為改觀。數據收集和運動表現的未來非常重要。智能傳感器的領域正在不斷擴大。傳感器可以與我們正在開發的 FlexWeave 技術相融合。FlexWeave 實際上是一種將材料編織在一起的方法，讓我們能夠將結構或耐久性結合到一個單一的包裝中，同時還能保持輕便和透氣。我們在實驗室裡記錄的那些東西，也可以在野外徒步記錄。是以，如果你在 FlexWeave 中加入了

Time for the Data Analyst's Dream Lair. Sweden's largest internet service provider is located somewhere you might not expect, below 100 feet of bedrock in a former nuclear bunker.

數據分析師的夢幻巢穴時間到了。瑞典最大的互聯網服務提供商位於一個你可能意想不到的地方，那就是位於 100 英尺深的基岩下的一個前核掩體中。

The Pionen Data Center houses some of the most sensitive information in the world, which once included WikiLeaks servers. They can help protect their clients' information under Sweden's Free Speech Act. Inside, engines intended to be for German submarines are used as backup generators. The design and style is the work of CEO John Carlin, who took inspiration from his love of films. He says that the choice to put plants in the caves was to capture the computer meets plants vibe of sci-fi films from the 70s. The plants are kept alive by grow lights. The center has simulated daylight, and the waterfall, well, that's just for fun. And while it may look like a villain's lair from James Bond, you won't find Goldfinger here. Just the 15 senior technical staff who run the servers.

Pionen 數據中心存放著世界上最敏感的資訊，其中曾經包括維基解密的服務器。根據瑞典的《言論自由法》，他們可以幫助保護客戶的資訊。在內部，原本用於德國潛艇的發動機被用作備用發電機。設計和風格是首席執行官約翰-卡林（John Carlin）的傑作，他從自己對電影的熱愛中汲取了靈感。他說，之所以選擇在洞穴中種植植物，是為了捕捉 70 年代科幻電影中電腦與植物的結合。這些植物通過生長燈來維持生命。中心有模擬日光，而瀑布只是為了好玩。雖然這裡看起來像詹姆斯-邦德中的反派巢穴，但你在這裡找不到金手指。只有運行服務