Can machines read your emotions? - Kostas Karpouzis

譯者: Songzhe Gao 審譯者: Cissy Yun

With every year, machines surpass humans in more and more activities

每年，以往被認為 只有人類能完成的事情

we once thought only we were capable of.

漸漸的被機械所超越

Today's computers can beat us in complex board games,

現在的電腦可以 在複雜的桌遊中打敗我們

transcribe speech in dozens of languages,

把演講翻譯成各種語言

and instantly identify almost any object.

還能立刻辨識各種物品

But the robots of tomorrow may go futher

在未來，機器人可能會透過

by learning to figure out what we're feeling.

理解我們的感受變得更進步

And why does that matter?

為什麼這很重要？

Because if machines and the people who run them

假如機器人以及操作者 可以精準判斷出我們的情感

can accurately read our emotional states,

they may be able to assist us or manipulate us

可能會給我們很大的幫助 或是進而操控我們

at unprecedented scales.

但在討論到那之前

But before we get there,

how can something so complex as emotion be converted into mere numbers,

試問為什麼像情緒這麼複雜的東西

可以轉換成機器唯一理解的 語言符號：數字呢？

the only language machines understand?

Essentially the same way our own brains interpret emotions,

其實跟大腦先學習認知情緒 才會詮釋自己的心情是一樣的

by learning how to spot them.

American psychologist Paul Ekman identified certain universal emotions

美國心理學家保羅·艾克曼 歸類出幾種廣泛存在的情緒

whose visual cues are understood the same way across cultures.

這些情緒的視覺線索 在各個文化中意義都相同

For example, an image of a smile signals joy to modern urban dwellers

舉例來說，一個笑臉圖像 對都市人和原住民來說

and aboriginal tribesmen alike.

都是代表喜悅

And according to Ekman,

根據艾克曼的說法

anger,

disgust,

憤怒

厭惡

fear,

恐懼

joy,

喜悅

sadness,

and surprise are equally recognizable.

悲傷

以及驚訝都同樣容易辨識

As it turns out, computers are rapidly getting better at image recognition

事實證明，電腦辨識 影像的速度愈來愈快

thanks to machine learning algorithms, such as neural networks.

多虧了像人工神經網絡 這種學習演算法

這些人工節點會藉由互相連結 和交換資訊來模仿生物神經的活動

These consist of artificial nodes that mimic our biological neurons

by forming connections and exchanging information.

To train the network, sample inputs pre-classified into different categories,

為了訓練網絡 會事先輸入預先歸類的樣本

such as photos marked happy or sad,

例如把標示快樂 或悲傷的照片輸入系統

are fed into the system.

The network then learns to classify those samples

網絡會根據特定的特徵調整相關數值 學習如何辨別這些樣本

by adjusting the relative weights assigned to particular features.

The more training data it's given,

輸入的訓練資料愈多 運算法辨識新資料就會愈準確

the better the algorithm becomes at correctly identifying new images.

This is similar to our own brains,

這跟我們的大腦一樣：

which learn from previous experiences to shape how new stimuli are processed.

藉由學習以往經驗 形成往後對新刺激的反應

Recognition algorithms aren't just limited to facial expressions.

辨識演算法不只會辨認臉部表情而已

Our emotions manifest in many ways.

情緒表達有很多種方式

There's body language and vocal tone,

包括肢體語言、語調

心率變化、膚色，以及表皮溫度

changes in heart rate, complexion, and skin temperature,

or even word frequency and sentence structure in our writing.

甚至包含說話節奏 和書面的語法結構

你可能認為訓練神經網絡 辨識它們是漫長又複雜的程序

You might think that training neural networks to recognize these

would be a long and complicated task

until you realize just how much data is out there,

不過當你了解網路數據的龐大 和現代電腦運算的速度就會改觀了

and how quickly modern computers can process it.

From social media posts,

像是從社群網站的發文 到上傳的照片和影片；

uploaded photos and videos,

and phone recordings,

從通話紀錄到熱感應監視器；

to heat-sensitive security cameras

and wearables that monitor physiological signs,

還有記錄生理狀況的穿戴式裝置

the big question is not how to collect enough data,

最大的問題已經不是如何取得資料

but what we're going to do with it.

而是使用這些資料的目的

There are plenty of beneficial uses for computerized emotion recognition.

電腦化表情辨識有很多實用途徑

Robots using algorithms to identify facial expressions

利用演算法辨識表情的機器人

can help children learn

可以幫助孩童學習

or provide lonely people with a sense of companionship.

或是陪伴孤單的人

Social media companies are considering using algorithms

社群媒體公司正在考慮運用演算法

to help prevent suicides by flagging posts that contain specific words or phrases.

來標記含有特定文字或用語的發文

以協助自殺防治

And emotion recognition software can help treat mental disorders

情緒辨識軟體 可以幫助治療心理疾病

or even provide people with low-cost automated psychotherapy.

甚至提供人們低價位的 自動化心理治療

Despite the potential benefits,

儘管有這些好處

the prospect of a massive network automatically scanning our photos,

大型網路自動掃描我們的照片

communications,

and physiological signs is also quite disturbing.

對話紀錄、和生理層面是相當惱人的

What are the implications for our privacy when such impersonal systems

當我們的情緒數據被公司用來打廣告 我們該如何保有隱私？

are used by corporations to exploit our emotions through advertising?

And what becomes of our rights

同時也出現人權問題

if authorities think they can identify the people likely to commit crimes

警方能夠把未決定犯罪的人 直接判定為罪犯嗎？

before they even make a conscious decision to act?

Robots currently have a long way to go

機器人的發展還有很長的路要走

例如辨別像諷刺這種細微的情緒

in distinguishing emotional nuances, like irony,

and scales of emotions, just how happy or sad someone is.

以及分辨情緒強弱 例如一個人多快樂或多難過

Nonetheless, they may eventually be able to accurately read our emotions

儘管如此，機器人將來 可能會精準地判斷情緒

and respond to them.

並做出回應

但是人類會不會因 機器人的過度入侵感到恐懼

Whether they can empathize with our fear of unwanted intrusion, however,

that's another story.

這又是另一回事了