Oh yeah, yeah. Yeah, yeah. Thank you.

喔對，對。謝謝

[MUSIC]

[音樂]

There have been a number of shifts in the way we think about computing

過去數十年來，我們對運算模式的思維

over the past few decades.

已有許多改變

The terminology artificial intelligence has come in and out of favor in the scientific community.

「人工智慧」這個術語在科學界爭議不斷

Sometimes it's called machine learning.

有時我們稱之為「機器學習」

We tend to call it machine intelligence these days.

最近傾向以「機器智能」稱呼

I just call it intelligence.

我都叫它「智能」

And sometimes it's just the effort to build machines that are better.

有時只是傾力打造更好的機器

So in the early days

在早期

everything was built on logic.

一切運算以邏輯式建立

Doing mathematical integration problems. Playing chess.

處理數學整合問題，下棋

But we realized that what the real challenges were were the things

但我們發現它真正的挑戰在於那些

that people can do every day.

人們每天都可以做到的小事情

The real world is actually very messy. Hard logical rules are not the way to solve

真實的世界是非常凌亂的，用複雜的邏輯運算也不能解決

really interesting real world problems.

這個有趣現實世界的問題

You have to have a system that will learn to get the knowledge in. You can't just

你必須有個會學習知識的系統，而不是

program it all in.

將所有問題寫進程式裡

Artificial intelligence is an effort to build machines that can learn from their environment,

人工智慧就是讓機器具有學習的能力，無論從環境、

from mistakes and from people.

錯誤中或從人身上學習皆可。

And we're still at the stage where we don't know what is the right path

我們仍在一個未知的階段，不曉得什麼才是正確的方向，

and the right breakthrough.

或是應有的突破

So I mean there's certainly a whole raft of different approaches.

但我想一定有許多不同的方法

One of the subfields we call pattern recognition.

其中一個分支稱為「模式識別」(用數學方法來研究生命體對環境及事物的自動處理及判讀，主要透過光學及聲學訊息識別)

Artificial neural network.

「人工神經網絡」(模仿生物神經的結構功能產生的計算模組)

Reinforcement learning, for example.

例如「強化學習」(機器透過觀察學習動作)

Statistical inference and probabilistic machine learning.

「統計推斷」和「概率機器學習」(統計中尋找參數最大值來估算未來行為，可說是用機率推算)

Supervised learning. Unsupervised learning. And we're not quite sure what technique is

「監督學習」「非監督學習」(經驗中產生函數，新數據使用所產生函數推算) 我們不確定哪種技術

going to lead to better systems. And, in fact, it's probably not one technique for everything,

能打造更好的系統。而且事實上，我們可能無法單靠一種技術解決所有事

it's probably a bunch of different techniques and combinations of those techniques.

而是同時採用各種技術及技術組合

Any progress we make in building truly intelligent systems is going to depend on progress in

我們在智能系統上的任何進展，大致上必須仰賴

technology generally.

科技的進步

And until recently, we didn't have computers that were fast enough or data sets that were

直到最近，我們才有夠快的電腦和夠大的數據機

big enough to do that.

足以做到這點

And so being able to take a particular problem and spread it out over lots and lots of machines

能夠將一個特定問題交由許多機台同時處理

is a very important approach because it makes our research faster.

是非常重大的進步，因為它縮短了研究時間。

So there's applications of artificial intelligence around us all the time.

我們周圍一直存在人工智慧的相關應用

When it begins to work or it does work it's all of a sudden given another name.

只是在工作初期或工作時被賦予了別的名稱

We're all already using it and very comfortable with it.

我們早就在使用，而且已經相當習慣了

Things that now we regard as routine 30 years ago would have been regarded as amazing

現在我們認為理所當然的事，在30年前被視為

examples of artificial intelligence.

人工智慧的驚人應用

Antilock braking.

防鎖死煞車系統

Autopilot systems for planes.

飛機自動駕駛

Search.

搜尋引擎

Recommendations.

推薦系統

Maps.

地圖

To decide whether or not this particular email is spam or not spam.

判別特定郵件是否為垃圾郵件

The ability to translate one language to another with your phone.

用手機翻譯成另一種語言的能力

Ten years ago if you tried to talk to your computer or to your phone, you know, that

十年前，如果你想和你的電腦或手機對話，你知道

would just be hopeless.

那是不可能的

We are seeing a steady torrent of these tricks one after the other getting figured out right now.

我們見證了一個個謎團，漸漸地水落石出

I think a lot of people that are close to the field have this

我想許多相關領域的人都有這種

do have that kind of breathless sense that things are moving quickly.

類似窒息的感覺，沒想到它發展得如此快速

It's a progressive thing. It's about building things that are slightly better

它是會成長的，就是創造某樣東西，它會自己一點點

slightly better, slightly better.

一點點，變得更好

Intelligence is really not going to be something that we ever succeed in defining in a succinct

智能真的不是像我們過去一樣，可以用簡明且

and singular way. It's really this whole constellation of different capabilities

單一的方式來定義。它涵蓋了千姿百態的能力範疇

that all kind of are beautifully orchestrated and working together.

而且都可以合作地天衣無縫

Predicting the long term future is very difficult.

預測日後的長遠發展是相當困難的

Nobody can really do it.

沒人能真正做到這點

And the bad thing to do is take whatever's working best now and assume the future's going to be like that forever.

但是我們不該預設，現在最先進的科技，在未來將一成不變

[MUSIC]

[音樂]