返回首页

HodLipson_2007-_“自我感知”机器人_
So, where are the robots? 啊,机器人们都在哪儿呢？
We've been told for 40 years already that they're coming soon. 四十年前人们就说它们很快就要来了
Very soon they'll be doing everything for us: they'll be cooking, cleaning, buying things, shopping, building. But they aren't here. 很快,它们就能为我们做一切事情 它们会做饭,打扫,买东西,购物,甚至是建房子。但直到今天,它们也没能进入我们的生活。
Meanwhile , we have illegal immigrants doing all the work, but we don't have any robots. 这段时间里,非法移民承担着这些工作, 但我们什么机器人都没有。
Meanwhile:adv.同时,其间;n.其间,其时; illegal:adj.不合法的;非法的;n.非法移民;非法劳工; immigrants:n.移民(immigrant的复数);
So what can we do about that? What can we say? 我们又能做些或说些什么呢？
So I want to give a little bit of a different perspective of how we can perhaps look at these things in a little bit of a different way. 我想给大家带来点不同的启发 看看今天我们能不能换一个角度看待这些事情
a little bit of a:一点点了; perspective:n.观点;远景;透视图;adj.透视的;
And this is an x-ray picture of a real beetle , and a Swiss watch, back from '88. You look at that -- what was true then is certainly true today. 这是一张X光片 上面有一只活甲虫,和一只88年的瑞士手表,你看—— 无论是当时还是现在
beetle:n.甲壳虫;adj.突出的;愁眉苦脸的;v.快速移动
We can still make the pieces, we can make the right pieces, we can make the circuitry of the right computational power, but we can't actually put them together to make something that will actually work and be as adaptive as these systems. 我们都可以做出这些零件,一模一样的零件, 我们能做出有同样计算能力的电路, 但我们并不能把它们放在一起,再做出个什么东西 能和这些系统（甲虫）有一样的适应力。
circuitry:n.电路;电路系统;电路学;一环路; computational:adj.计算的; adaptive:adj.适应的,适合的;
So let's try to look at it from a different perspective. 那么就让我们试着换个角度再看看这个问题。
Let's summon the best designer, the mother of all designers: let's see what evolution can do for us. 让我们召集最好的设计师,所有设计师的鼻祖, 看看进化论能给我们做些什么。
summon:v.召唤;召集;鼓起;振作; evolution:n.演变;进化;发展;渐进;
So we threw in -- we created a primordial soup with lots of pieces of robots: with bars, with motors, with neurons. 所以我们就找来了--我们创造了一锅“原汤”, 里面有组装机器人需要的各种零件:条状的,带马达的,还有些带神经元的。
primordial:adj.原始的;根本的;原生的;
Put them all together, and put all this under kind of natural selection , under mutation , and rewarded things for how well they can move forward. 把它们都放在一起,然后把所有这些放到一种自然选择, 可能产生突变的环境中,并奖励那些发生“进化”的零件组。
natural selection:n.自然选择;物竞天择; mutation:n.[遗]突变;变化;元音变化; rewarded:v.奖励;奖赏;给以报酬;(reward的过去分词和过去式)
A very simple task, and it's interesting to see what kind of things came out of that. 这是一个非常简单的任务,并且得出的结果非常有趣。
So if you look, you can see a lot of different machines come out of this. They all move around , they all crawl in different ways, and you can see on the right, that we actually made a couple of these things, 看一下,你能看到各式各样的机器 在这种模式下制造出来,它们到处移动, 以不同的方式爬行,在右边你可以看到 我们真的做出了几个这样的玩意儿,
move around:v.走来走去;绕着…来回转; crawl:v.爬;匍匐行进;(昆虫)爬行;缓慢行进;n.爬泳;缓慢的速度;自由泳;
and they work in reality. These are not very fantastic robots, but they evolved to do exactly what we reward them for: for moving forward. So that was all done in simulation , but we can also do that on a real machine. 它们在现实中真的能工作。这些还算不上非常先进的机器人, 但它们的确按照我们奖励的方向进化了: 那就是向前进化。这些都是模拟的, 但我们在真的机器上也做成功了,
fantastic:奇异的,空想的 evolved:v.(使)逐渐形成;进化;进化形成;(evolve的过去分词和过去式) simulation:n.仿真;模拟;模仿;假装;
Here's a physical robot that we actually have a population of brains, competing , or evolving , on the machine. 这是一个真实的机器人, 我们动用了大量的人力脑力 让机器们相互竞争,共同进化
physical:adj.[物]物理的;身体的;物质的;符合自然法则的;n.体格检查; competing:adj.相互冲突的;相互矛盾的;v.竞争;对抗;参加比赛;(compete的现在分词) evolving:v.(使)逐渐形成,逐渐演变;进化形成;(evolve的现在分词)
It's like a rodeo show: they all get a ride on the machine, and they get rewarded for how fast or how far they can make the machine move forward. 这有点像一场套马表演:他们都骑在机器上 根据他们能够驾驶机器前进的速度和距离 得到奖励
rodeo:n.竞技表演;竞技者;vi.竞技;
And you can see these robots are not ready to take over the world yet, but they gradually learn how to move forward, and they do this autonomously . 你可以看到这些机器人都还没完全准备好 占领这个世界,但是 它们渐渐地学会如何前进 并且是自发地学习。
take over:接管;继承;接收;接任;接替; gradually:adv.渐渐地;逐步地; autonomously:adv.自治地;独立自主地;
So in these two examples, we had basically machines that learned how to walk in simulation, and also machines that learned how to walk in reality. 所以从这两个例子中,我们已经基本上 得到了能在虚拟中学习走路 和在现实中学习走路的机器人。
basically:adv.主要地,基本上;
But I want to show you a different approach , and this is this robot, over here, which has four legs, it has eight motors, four on the knees and four on the hip . 现在我还要再给你们展示另一项进展, 就是这里的这个机器人,在这儿,它有四条腿 它身上安了八个马达,膝盖上四个,腿上四个
approach:n.方法;路径;v.接近;建议;着手处理; hip:n.臀部;蔷薇果;忧郁;adj.熟悉内情的;非常时尚的;
It has also two tilt sensors that tell the machine which way it's tilting . 它还配备了两个倾斜度传感器,可以告诉自己 正在向哪个方向倾斜。
sensors:n.[自]传感器,感应器;感测器(sensor的复数); tilting:n.倾卸台;adj.[航][气象]倾斜;倾卸;v.使倾斜(tilt的ing形式);
But this machine doesn't know what it looks like. 但这个机器并不知道自己长啥样
You look at it and you see it has four legs, the machine doesn't know if it's a snake, if it's a tree, it doesn't have any idea what it looks like, but it's going to try to find that out. 你可以看到它长了四条腿, 但它自己并不知道自己是一条蛇还是一棵树 它完全给蒙在鼓里,不晓得自己的相貌 但它马上就要试着找到自己
Initially , it does some random motion , and then it tries to figure out what it might look like -- and you're seeing a lot of things passing through its minds, 首先,它会做一些随机的动作 然后试着弄清楚那些动作都看起来是什么样子的—— 你可以看到它的脑海里闪过许许多多的东西,
Initially:adv.最初,首先;开头; random:adj.[数]随机的;任意的;胡乱的;n.随意;adv.胡乱地; motion:n.动作;移动;手势;请求;意向;议案;v.运动;打手势;
a lot of self-models that try to explain the relationship between actuation and sensing -- and then tries to do a second action that creates the most disagreement among predictions of these alternative models, like a scientist in a lab. Then it does that and tries to explain that, and prune out its self-models. 大量的自我尝试的动作模型,试着理清 行动和感官之间的关系——然后它将再做第二个动作 在所有可能的动作模型中 最诡异的一个动作, 就像科学家在实验室里的试验。接着,它重复那个动作, 并且试着解释那个动作,然后梳理出自己的动作模型。
actuation:n.冲动,驱使;刺激;行动; predictions:n.预测,预言(prediction复数形式); alternative:adj.供选择的;选择性的;交替的;n.二中择一;供替代的选择; prune:vi.删除;减少;vt.修剪;删除;剪去;n.深紫红色;傻瓜;李子干;
This is the last cycle, and you can see it's pretty much figured out what its self looks like, and once it has a self-model, it can use that to derive a pattern of locomotion . 这是最后一个环节,你可以看到它已经基本上 清楚自己的样子了。一旦它理清自己的动作模型, 就可以从模型中得出一种运动模式。
derive:vt.源于;得自;vi.起源; locomotion:n.运动;移动;旅行;
So what you're seeing here are a couple of machines -- a pattern of locomotion. 好了,你现在看到的是几个机器—— 嗯,一种运动模式,
We were hoping that it's going to have a kind of evil , spidery walk, but instead, it created this pretty lame way of moving forward. 我们期待它能产生一种邪恶的,蜘蛛式的运动, 结果它却自创出这种相当脑残的前进方式。
evil:adj.邪恶的;不幸的;有害的;讨厌的;n.罪恶,邪恶;不幸; spidery:adj.细长足的;蜘蛛一般的;蜘蛛网一般的; lame:adj.瘸的; n.(金属)薄板; v.使残废;
But when you look at that, you have to remember that this machine did not do any physical trials on how to move forward, nor did it have a model of itself. 但当你看着它前进的时候,你必须记得 这个机器并没有接受任何物理指令,控制着它们向前进, 它也没有任何已有的自我模型。
It kind of figured out what it looks like, and how to move forward, and then actually tried that out. 它相当于是自己推理出了自己的样子,以及应该如何向前进 并且进行了亲身的尝试。
(Applause) 鼓掌~~~?
So, we'll move forward to a different idea. 那么现在,我们再来看看另一个想法。
So that was what happened when we had a couple of -- that's what happened when you had a couple of -- OK, OK, OK -- 那是我们将几个... 把几个...放在一块儿就会...好啦好啦好啦——
(Laughter) （笑）
- they don't like each other. So there's a different robot. ——它们不大喜欢对方,所以啦~ 这是另外一个机器人。
That's what happened when the robots actually are rewarded for doing something. 刚才的那些都是在机器人做对了动作, 获得奖励的情况下发生的。
What happens if you don't reward them for anything, you just throw them in? 那么如果我们不给它们奖励,只是把它们扔到一块,又会怎么样呢？
So we have these cubes , like the diagram showed here. 所以我们拿来了这些立方体,就像这里的这些图,
cubes:n.[数]立方体,数据库(cube的复数形式); v.使成立方形(cube的第三人称单数形式);
The cube can swivel , or flip on its side, and we just throw 1,000 of these cubes into a soup -- this is in simulation --and don't reward them for anything, we just let them flip. We pump energy into this and see what happens in a couple of mutations . 它们能旋转,或者翻筋斗 我们把1000个这样的立方体放入“原汤”—— 这是模拟效果——我们没给它们任何奖励, 我们就让它们自己活动。我们给它们注入了些能量, 看看经过几次突变,会发生点什么。
swivel:n.转环;转椅座架;vt.使旋转;vi.旋转; flip:n.浏览;空翻;轻抛;捻掷;v.迅速翻动;按(开关);按(按钮);开(或关)(机器等); pump:n.抽水机;泵;打气筒;v.用泵输送;涌出;奔流;快速摇动; mutations:n.[遗]突变;变化;转变(mutation的复数形式);
So initially, nothing happens, they're just flipping around there. 刚开始的时候,什么也没发生,它们光在那儿跳来跳去。
flipping:v.该死; v.(使)快速翻转; (flip的现在分词)
But after a very short while, you can see these blue things on the right there begin to take over. 但过了一小会儿,你可以看到这些蓝色的小东西, 它们在右边逐渐地开始占取主动。
They begin to self-replicate. So in absence of any reward, the intrinsic reward is self-replication. 它们开始自我复制。由此可见即使没有任何奖励 它们也会用自我复制的方式来奖励自己。
absence:n.没有;缺乏;缺席;不注意; intrinsic:adj.本质的,固有的;
And we've actually built a couple of these, and this is part of a larger robot made out of these cubes, it's an accelerated view, where you can see the robot actually carrying out some of its replication process . 事实上我们已经造了好几个这类的玩意儿, 这是一部分以这些立方体为单位造出来的大机器人, 这是快进的效果,可以让你看到机器人 进行自我复制的过程。
accelerated:v.(使)加速;加速;加快(accelerate的过去分词和过去式) replication:n.复制;回答;反响; process:v.处理;加工;列队行进;n.过程,进行;方法,adj.经过特殊加工(或处理)的;
So you're feeding it with more material -- cubes in this case -- and more energy, and it can make another robot. 如果你给它多喂点儿——就是这些立方体—— 再多给它点能量,它就能自己造出另一个机器人。
So of course, this is a very crude machine, but we're working on a micro-scale version of these, and hopefully the cubes will be like a powder that you pour in. 当然,这还是一个非常粗糙,不成熟的机器, 但我们正在研究微缩版的这类机器人, 希望这些立方体能小到像倒出的面粉一般。
crude:adj.粗略的;简略的;大概的;粗糙的;n.原油;石油; micro-scale:微尺度; powder:n.粉末;细面;扑面粉;美容粉;v.傅粉;抹粉; pour:v.倒; n.流出; (已熔金属的)一次浇注量;
OK, so what can we learn? These robots are of course not very useful in themselves, but they might teach us something about how we can build better robots, and perhaps how humans, animals, create self-models and learn. 好的,那么我们都了解到了什么？这些机器人当然 自己并不是有很大用处,但它们能教会我们一些东西, 让我们知道如何造出更好的机器人, 甚至是人类,动物创造自我模型和学习机制的原理。
And one of the things that I think is important is that we have to get away from this idea of designing the machines manually , but actually let them evolve and learn, like children, and perhaps that's the way we'll get there. Thank you. 我觉得这其中最重要的, 就是我们必须摒弃之前的观念, 手动地设计这些机器 而是让它们自己进化,学习,像孩子一样, 这大概才是我们成功的必经之路。谢谢！
manually:adv.手动地;用手;
(Applause) （鼓掌?）