跟读练习: Turing & The Halting Problem - Computerphile - 通过视频学习英语口语

Today we're going to be talking about a problem in logic, and how in solving that problem,

Alan Turing almost inadvertently invented the modern digital computer.

So we start back at the beginning of the 20th century, where mathematicians had posed this problem.

- Thank you.

In logic, we're interested in finding, do these premises entail this conclusion?

So premises are the bits you start off with in an argument.

They are the bits you know at the beginning or your assumptions.

And the conclusion is the bit you want to establish, the bit that you reason to with your argument.

And we want to know, is there a test that will tell us, yes, for sure, these premises do or don't entail this conclusion?

Is there an automatic way of finding out whether they do or whether they don't. So that's the problem.

It's called the decision problem.

and mathematicians wanted to find out Is there an answer to the decision problem for first order logic?

That's the kind of logic you learn in philosophy or mathematics at university.

So lots of mathematicians were trying to work out, is first order logic decidable?

That is, can we automatically test whether the premises entail the conclusion?

Alan Shuring was one of the first to discover that first order logic isn't decidable.

I To prove this, It's really difficult conceptually,

because you have to be able to show no possible program can give you the answer.

But how do you do that?

How do you show something about every possible program?

You can't run through every program one by one.

But Turing came up with a brilliant solution.

Yeah.

His idea goes something like this.

Suppose we have a program and let's just draw it as a black box.

It's going to take some inputs, and it's going to give us some outputs.

our program is going to solve some problem.

a problem like, do the premises entail the conclusion?

We ask it a question and it will give us an answer: yes, or no?

Now here's another question we can ask: Thank you.

Let's look at all of those possible programs.

And we're just thinking of them as black boxes at the moment.

We might want to know, is this program Given a certain input, gonna give us an answer.

Or is it going to trundle on forever and never give us an answer?

That is, is it going to halt or is it not going to halt eventually?

So think about your computer running.

You want it to give you an answer.

And whether it's a good answer or a bad answer, it's better than no answer.

No answer would mean the computer trundles round forever and ever in a loop.

And you would just never know whether it's going to finish today, tomorrow or never.

So halting is good.

So there's another question we can ask: Given some program and some inputs, will it ever halt?

Now it turns out that our logical problem Do these premises entail this conclusion?

is very similar to this halting problem.

In fact, if we can solve the logical problem, then we can solve the halting problem.

Will this program halt on this input?

So the clever part of Turing's proof is to show that it's impossible for any machine, however clever it is, to solve the halting problem.

That is to tell us whether a given machine with a given input will halt or not.

And here's how we did it.

Let's suppose we've got a machine or a program that solves the problem for us it solves the holding problem.

Don't worry about how it works, let's just think of it as a black box, taking a description of a machine and an input, and giving us an answer.

Yes it will halt or no it won't halt.

Just suppose that's possible.

Call that machine H for the halting problem.

If you give me that machine, I can transform it into a different machine like this.

I stick some extra bits on it so that If it gives me a yes answer, I make it loop forever without ever stopping.

If it gives me a no answer on the other hand, it's gonna halt straight away.

Let's call that big machine the whole thing.

H plus.

Now here's another question we can ask: what happens if I feed the whole machine into itself.

So I'm going to put H plus Thank you in here and H + in here.

So the question I'm now asking is, I'm feeding H + into itself.

So I'm asking the question: Does H + halt?

give an input H+.

And here's where it all goes wrong.

because If H + does halt, We get a yes answer.

But then it loops forever, so it doesn't halt.

On the other hand, if it doesn't halt, We get a no answer, but then it halts.

So if it does halt, then it doesn't halt.

But if it doesn't halt, then it does halt.

Either way, we get a contradiction.

It's a paradox.

And what that shows is we started off assuming that we can solve the problem.

we've ended up with a paradox.

So our assumption was bad.

It turns out there's no possible machine, no possible program that solves the halting problem.

The really clever bit about Turing's idea is it doesn't matter what kind of program or machine it is.

It doesn't matter whether it's An abstract algorithm, whether it's a real computer, a physical computer, it doesn't matter what it is.

We've proved that no such program is possible.

you Turing, as part of his argument, had to say a little bit about what's going on in these black boxes.

And the idea is that every card represents an instruction in this Turing machine.

📺 同频道

Why AI Tokens are so Expensive - Computerphile

Business EnglishUniversity StudentsWorking Adults

✨ 推荐视频

182 句

本课的词汇与口语要点

这段视频共有 81 个句子、976 个单词可供跟读。讲话部分时长为 6:14。说话人语速自然，每分钟约 157 个词，接近日常对话。 83% 的单词属于英语最常用的 3,000 词，其余的词建议在练习前先学一下。

视频中的重点词汇

视频中 15 个值得学习的单词，附发音和释义：

单词	发音	释义
hungry 形容词	/ˈhʌŋ.ɡɹi/	飢餓 /饥饿, 餓 /饿
solve 动词	/sɒlv/	解決 /解决
input 名词	/ˈɪnpʊt/	輸入 /输入
conclusion 名词	/kənˈkluːʒən/	結束 /结束
premise 名词	/ˈpɹɛm.ɪs/	前提
loop 名词	/luːp/	圈, 環 /环
assumption 名词	/əˈsʌm(p).ʃ(ə)n/	假設 /假设
paradox 名词	/ˈpæ.ɹəˌdɒks/	悖論 /悖论, 矛盾
establish 动词	/ɪˈstæb.lɪʃ/	確定 /确定
output 名词	/ˈaʊtpʊt/	輸出 /输出
discover 动词	/dɪˈskʌvə/	發現 /发现, 發覺 /发觉
automatic 形容词	/ˌɔː.təˈmæt.ɪk/	自動的 /自动的
mathematics 名词	/mæθ(.ə)ˈmæt.ɪks/	數學 /数学, 算學 /算学
abstract 名词	/ˈæbˌstɹækt/	摘要
instruction 名词	/ɪnˈstɹʌkʃən/	教導 /教导

视频中出现的短语动词

单词	发音	释义
find out 动词		找出, 查明
turn out 动词		原來 /原来
come up with 动词		想出, 提出
work out 动词	/ˌwɝk ˈaʊt/	計算 /计算, 算出

需要注意的发音

说话人用了 25 次缩略和弱读形式，例如 doesn't, I'm, don't。请按听到的简短形式来说。

“th” 音: mathematics /mæθ(.ə)ˈmæt.ɪks/, algorithm /ˈælɡəɹɪðm̩/, mathematician /ˌmæθ.(ə.)məˈtɪʃ.ən/
“sh” 和 “zh” 音: conclusion /kənˈkluːʒən/, assumption /əˈsʌm(p).ʃ(ə)n/, establish /ɪˈstæb.lɪʃ/, instruction /ɪnˈstɹʌkʃən/, mathematician /ˌmæθ.(ə.)məˈtɪʃ.ən/
长单词——注意重音位置: automatic /ˌɔː.təˈmæt.ɪk/, mathematics /mæθ(.ə)ˈmæt.ɪks/, mathematician /ˌmæθ.(ə.)məˈtɪʃ.ən/, decidable [dɪˈsaɪdəbəɫ], conceptually /kənˈsɛp.tjʊə.li/

如何用这段视频练习

先完整听一遍视频，不要开口，记下不认识的单词。
先用 0.75 倍速逐句跟读，熟练之后再回到正常速度。
录下自己的声音并与原声对比，特别注意 hungry, solve, input 这类单词。

什么是跟读法？

跟读法 (Shadowing) 是一种有科学依据的语言学习技巧，最初开发用于专业口译员的培训，并由多语言者Alexander Arguelles博士普及。这个方法简单而强大：您在听英语母语原声的同时立即大声重复——就像是一个延迟1-2秒紧跟说话者的影子。与被动听力或语法练习不同，跟读法强迫您的大脑和口腔肌肉同时处理并模仿真实的讲话模式。研究表明它能显着提高发音准确性，语调，节奏，连读，听力理解和口语流利度——使其成为雅思口语备考和真实英语交流最有效的方法之一。

影子跟读法: 阅读完整分步指南 →

跟读练习: Turing & The Halting Problem - Computerphile - 通过视频学习英语口语

Why AI Tokens are so Expensive - Computerphile

We Let $200 GPT-Pro Think for 7 Days on Unsolved Math

What is a PLC and how does it work?

🚫 Java Deadlock Explained: Why Your Code is Stuck 😱 (Beginners Must Watch!)

CAP Theorem Simplified

Physicist Brian Cox explains quantum physics in 22 minutes

Traceroute: You're Probably Reading the Results Wrong

CS50x 2026 - Introduction

[C1] Stop Overthinking | English Listening & Speaking Practice

Mathematician Collapses All Functions to One Weird Formula

AI biết code, vì sao vẫn phải học lập trình? | đi-code ep 35

Force Your Brain to Think Like a Genius | Richard Feynman

LangChain in 5 Minutes (Explained Clearly)