but the truth is it's nothing like the brains found in nature or in our skulls.

但事實是，它完全不像在自然界或我們的頭骨中發現的大腦。

CPUs are great at performing precise calculations with huge numbers,

CPU很擅長用龐大的數字進行精確計算。

but when it comes to learning and abstraction, the thinky meat between our ears has the CPU licked.

但在學習和抽象的時候，我們耳朵間的那塊思念的肉卻讓CPU給舔了。

An emerging field of artificial intelligence called neuromorphic computing is attempting to mimic how the neurons in our own brains work,

一個被稱為神經形態計算的新興人工智能領域正試圖模仿我們自己大腦中的神經元如何工作。

and researchers from Intel and IBM are making true silicon brains a reality.

和來自英特爾和IBM的研究人員正在使真正的硅腦成為現實。

Now, it's easy to get a little lost in the terminology here because another technology on the forefront of AI is called deep learning,

現在，這裡的術語很容易讓人有點迷茫，因為人工智能最前沿的另一項技術叫做深度學習。

and one of the most advanced approaches relies on something called a neural network.

而最先進的方法之一就是依靠一種叫做神經網絡的東西。

Neural networks are a software approach that mimic how brains work.

神經網絡是一種模擬大腦工作方式的軟件方法。

A neural network changes when it's shown lots and lots of examples of what it's supposed to learn,

當一個神經網絡被展示了很多很多的例子時，它應該學習的東西就會改變。

but it may need to see thousands to millions of examples to achieve the desired results,

但它可能需要看到幾千到幾百萬的例子才能達到預期的效果。

like how to tell the difference between a chihuahua and a blueberry muffin.

比如如何分辨吉娃娃和藍莓鬆餅的區別。

Clearly that's not how we learn.

顯然，這不是我們的學習方式。

I don't need to see millions of pictures of a dog before I know what a dog is.

我不需要看到幾百萬張狗的照片，就知道什麼是狗。

But if you want to send me pictures of your dog I am on Twitter.

但如果你想把你的狗的照片發給我，我在微博上。

So, to solve this, researchers from IBM and Intel are trying to mimic brains at a hardware level too.

所以，為了解決這個問題，來自IBM和英特爾的研究人員也在嘗試在硬件層面模仿大腦。

IBM revealed their brain inspired chip called TrueNorth in 2014,

IBM在2014年揭示了他們的大腦啟發芯片，名為TrueNorth。

while Intel's chip called Loihi was introduced in 2017.

而英特爾在2017年推出了名為Loihi的芯片。

The two neuromorphic chips use the same silicon transistors commonly found in conventional chips,

這兩款神經形態芯片使用的是傳統芯片中常見的硅半導體。

but they're arranged to interconnect more like neurons.

但它們的排列方式更像神經元的互連。

TrueNorth's one million neurons are connected by 256 million synapses,

TrueNorth的100萬個神經元由2.56億個突觸連接。

while Loihi's 130,000 neurons are each capable of communicating with thousands of others for a total of over 130 million synapses.

而洛伊希的13萬個神經元，每個神經元都能與其他數千個神經元進行交流，總計超過1.3億個突觸。

TrueNorth and Loihi also combined into one chip two aspects of computers that are normally separate: memory and computation.

TrueNorth和Loihi還將計算機通常分離的兩個方面結合到一個芯片中：內存和計算。

In a typical computer like you have at home, the CPU handles computation and shuffles data back and forth from the Random Access Memory, or RAM.

在典型的電腦中，比如你家裡的電腦，CPU負責處理計算，並將數據從隨機存取存儲器或RAM中來回洗牌。

But this separation slows things down and draws more power, and it's not how things work in our own brains.

但這種分離會使事情的發展速度變慢，汲取更多的力量，這不是我們自己大腦中的工作方式。

In another drastic departure from standard chips, TrueNorth and Loihi do not use a clock to update information across the system in a synchronized manner.

在另一個與標準芯片的巨大差異中，TrueNorth和Loihi不使用時鐘來同步更新整個系統的資訊。

Instead, the neurons in the chip fire independently, and the timing of these spikes of activity can be used as another way to encode information.

相反，芯片中的神經元是獨立發射的，這些活動尖峰的時間可以作為編碼資訊的另一種方式。

All of these tweaks to how information is moved around means neuromorphic chips can learn quickly and use far less energy than a conventional CPU.

所有這些對資訊移動方式的調整都意味著神經形態芯片可以快速學習，而且能耗遠遠低於傳統CPU。

Best of all though, is the problems they can solve as a result of their novel design.

不過最好的是，他們的新穎設計所能解決的問題。

Problems like constraint satisfaction, where several solutions could exist but only one of them fits the constraints.

像約束條件滿足這樣的問題，可能存在幾個解，但其中只有一個符合約束條件。

Think Sudoku puzzles.

想想數獨謎題。

Neuromorphic computers can also be used for optimization tasks, like the famous traveling salesman problem

神經形態計算機也可以用於優化任務，比如著名的旅行推銷員問題。

where finding the best route to take from millions of options can be very challenging, even for a supercomputer.

在這裡，從數以百萬計的選項中尋找最佳路線是非常具有挑戰性的，即使是對一臺超級計算機來說也是如此。

Since Loihi is a research chip that was never intended for mass production, there aren't many of them for researchers to work with.

由於Loihi是研究芯片，從來沒有打算量產，所以供研究人員使用的芯片並不多。

Still, Intel wired together 768 of them to create Pohoiki Springs, a computer that's the size of 5 servers and boasts 100 million neurons.

不過，英特爾還是將其中的768個連接在一起，創造出了Pohoiki Springs，這臺電腦的大小相當於5臺服務器，擁有1億個神經元。

That's in league with the brain size of a small mammal.

這與小型哺乳動物的大腦大小是一致的。

And yet, despite its size and complexity, it needed under 500 watts of power to operate.

然而，儘管它的規模和複雜性，它需要500瓦以下的功率才能運行。

By contrast, the overkill gaming PC sitting next to me can use up to twice that much power, and it still isn't as “smart” as a squirrel.

相比之下，坐在我旁邊的那臺矯枉過正的遊戲電腦，可以用上兩倍的電量，而且還不如一隻松鼠 "聰明"。

Neuromorphic computers are not poised to completely replace conventional ones any time soon.

神經形態計算機還沒有準備好在短時間內完全取代傳統計算機。

Remember that because this kind of hardware is just emerging, software that can make the best use of it needs time to develop.

請記住，因為這種硬件剛剛興起，所以能夠充分利用它的軟件需要時間來開發。

Still, it's something to look forward to.

不過，這還是值得期待的。

As the technology matures we'll be able to crack bigger and tougher problems that were previously beyond our grasp with our current CPU "brains."

隨著技術的成熟，我們將能夠破解更大、更難的問題，而這些問題是以前我們目前的CPU "大腦 "無法掌握的。

While our brains are more adaptable than a conventional CPU,

雖然我們的大腦比傳統CPU的適應性更強。

our data processing speed is estimated to be a paltry 120 bits per second.

我們的數據處理速度估計只有每秒120比特。

If you want to know more about neural networks, check out Maren's video on how robots teach themselves here!

如果你想了解更多關於神經網絡的知識，可以在這裡查看Maren關於機器人如何自學的視頻!

If you like this video be sure to let us know in the comments, or subscribe!

如果你喜歡這個視頻，一定要在評論中告訴我們，或者訂閱!

Then we know you really like us.

那我們就知道你是真的喜歡我們。

Thanks for watching and I'll see you next time on Seeker!

謝謝你的觀看，我們下期《尋人啟事》再見!