ADM-201 dump PMP dumps pdf SSCP exam materials CBAP exam sample questions

机器能否读心 – 译学馆
未登陆,请登陆后再发表信息
最新评论 (0)
播放视频

机器能否读心

Can machines read your emotions? - Kostas Karpouzis

【启点字幕组】
Light up the world
随着时代发展 机器越来越多地超越人类
With every year, machines surpass humans in more and more activities
完成各种我们曾以为只有人类能做到的事
we once thought only we were capable of.
如今的电脑能在各种复杂的桌游中战胜人类
Today’s computers can beat us in complex board games,
将文字翻译成各种语言
transcribe speech in dozens of languages,
即刻辨识物体
and instantly identify almost any object.
而未来的机器将会更进一步
But the robots of tomorrow may go futher
学会解读人类的情感
by learning to figure out what we’re feeling.
为什么这很重要呢
And why does that matter?
因为如果机器及其操控者
Because if machines and the people who run them
能够准确解读我们的情绪状态
can accurately read our emotional states,
他们就可能以前所未有的形式和深度
they may be able to assist us or manipulate us
帮助或者操纵我们
at unprecedented scales.
但在此之前
But before we get there,
复杂的情绪如何能转换成数字呢
how can something so complex as emotion be converted into mere numbers,
毕竟数字是计算机唯一的语言
the only language machines understand?
本质上 机器可以像人脑一样
Essentially the same way our own brains interpret emotions,
学会识别进而解读情绪
by learning how to spot them.
美国心理学家保罗·艾克曼发现了一些普遍的情绪
American psychologist Paul Ekman identified certain universal emotions
不论文化背景如何 这些情绪的表现形式都大同小异
whose visual cues are understood the same way across cultures.
例如 不论是现代城市居民 还是土著部落
For example, an image of a smile signals joy to modern urban dwellers
微笑都代表快乐
and aboriginal tribesmen alike.
根据艾克曼的研究
And according to Ekman,
愤怒
anger,
厌恶
disgust,
恐惧
fear,
快乐
joy,
悲伤
sadness,
以及惊讶 都普遍易于识别
and surprise are equally recognizable.
事实证明 电脑正在快速优化图像识别系统
As it turns out, computers are rapidly getting better at image recognition
这要归功于机器学习算法 比如模拟神经网络
thanks to machine learning algorithms, such as neural networks.
用模拟生物神经元的人工节点
These consist of artificial nodes that mimic our biological neurons
来建立连接 交换信息
by forming connections and exchanging information.
然后训练网络 输入样本时 将样本预分至不同类别下
To train the network, sample inputs pre-classified into different categories,
如区分高兴和悲伤的照片
such as photos marked happy or sad,
将分类样本输入系统
are fed into the system.
网络通过调整对特定特征的相对权重
The network then learns to classify those samples
学习对样本进行区分
by adjusting the relative weights assigned to particular features.
训练数据积累得越多
The more training data it’s given,
算法对新图片的识别就越准确
the better the algorithm becomes at correctly identifying new images.
这与我们大脑的处理方式相似
This is similar to our own brains,
大脑也是依据过往经验学习如何处理新刺激
which learn from previous experiences to shape how new stimuli are processed.
识别算法不仅局限于面部表情
Recognition algorithms aren’t just limited to facial expressions.
情绪有多种表现方式
Our emotions manifest in many ways.
比如 肢体动作 语音语调
There’s body language and vocal tone,
心率变化 脸色 体表温度
changes in heart rate, complexion, and skin temperature,
甚至是写作时的用词频率 句式结构
or even word frequency and sentence structure in our writing.
你可能觉得 要机器学会识别上述内容
You might think that training neural networks to recognize these
将是一个漫长而复杂的的过程
would be a long and complicated task
然而 我们拥有大量数据
until you realize just how much data is out there,
现代电脑可以很快处理这些数据
and how quickly modern computers can process it.
从社交媒体发文
>From social media posts,
人们上传的照片 视频
uploaded photos and videos,
电话录音
and phone recordings,
到监控录像
to heat-sensitive security cameras
以及可监测生理数据的可穿戴设备
and wearables that monitor physiological signs,
我们不必担心数据不够
the big question is not how to collect enough data,
而应考虑如何利用这么大量的数据
but what we’re going to do with it.
运用计算机识别情绪能带来许多益处
There are plenty of beneficial uses for computerized emotion recognition.
机器人运用算法识别面部表情
Robots using algorithms to identify facial expressions
可以帮助孩子们学习
can help children learn
或者为孤独的人提供陪伴
or provide lonely people with a sense of companionship.
社交媒体公司正在考虑使用这个算法
Social media companies are considering using algorithms
来识别特定关键词以预防自杀事件
to help prevent suicides by flagging posts that contain specific words or phrases.
情绪识别软件还有助于治疗精神疾病
And emotion recognition software can help treat mental disorders
甚至为人们提供低成本的自动化治疗方案
or even provide people with low-cost automated psychotherapy.
虽然好处不少
Despite the potential benefits,
但一想到网络将自动监测我们的照片
the prospect of a massive network automatically scanning our photos,
通话
communications,
和生理状况 就难免令人感到不安
and physiological signs is also quite disturbing.
如果企业在广告宣传中使用这类系统操控我们的情绪
What are the implications for our privacy when such impersonal systems
那么我们的个人隐私将如何得到保障
are used by corporations to exploit our emotions through advertising?
如果政府使用这类系统
And what becomes of our rights
将尚未产生犯罪意图的人判定为潜在罪犯
if authorities think they can identify the people likely to commit crimes
那我们的权利何在
before they even make a conscious decision to act?
机器读心技术还有很长的路要走
Robots currently have a long way to go
如何区分情绪上的细微差别 如何识别讽刺
in distinguishing emotional nuances, like irony,
如何判定感情的强烈程度 如何确定有多开心或有多悲伤
and scales of emotions, just how happy or sad someone is.
尽管如此 机器终有一天能做到准确解读人类情绪
Nonetheless, they may eventually be able to accurately read our emotions
并对人类情绪做出反应
and respond to them.
但是机器能否理解我们对其来之不速的恐慌呢
Whether they can empathize with our fear of unwanted intrusion, however,
那可能就要另当别论了
that’s another story.

发表评论

译制信息
视频概述
听录译者

收集自网络

翻译译者

启点-洞洞

审核员

自动通过审核

视频来源

https://www.youtube.com/watch?v=QFk3e5PcK7s

相关推荐