Skip to content

The future was born here… You will look back and see it I think… Jeff Hawkins 2012 ICSA Speech

December 8, 2012

THE MONUMENTAL  significance of Jeff Hawkins’ lifetime of learning about neuroscience and integrating it with computer science (as one of the FATHERS of mobile computing  – inventor and founder of Palm and Handspring) – this CONVERGENCE – cannot be overstated or overemphasized.

I could be completely wrong – and t/his company, Numenta, along with his ‘theories’ (even though they are now being implemented in actual working machines which you can SEE) could somehow flop because something unpredictable happens.

That kind of thing does happen, and is possible in life, of course.

So – I can only say what I THINK will happen.

I hope Mr. Hawkins would like my tribute  and homage to him here, and my initial attempt to share my understanding of what he is saying and has said (in his book and afterwards in many times and places). 

If I err, it is because my brain is an imperfect (for certain) hierarchical memory-prediction system.  My brain still needs serious ‘training,’ and I’m trying to train it – even though I did start at a pretty young age 🙂 

I am training it now in the IT field and cybersecurity, but I do have Gigabytes of partially useful information and I’m perpetually fascinated with SO many areas of life – but especially brain science and cognitive psychology and now this convergence.

I think that this convergence, and his company called Numenta (or whatever it morphs into by the time it becomes ‘well known’) will create the biggest leap forward in so many disciplines that it is hard to list. Perhaps it on par with Einstein’s Theory of Relativity (but for the brain and intelligence).

After all, he is/was friends with Watson and Crick (yes, that Watson and that Crick of DNA fame) – or whichever one of them is around these days. I don’t call exactly.

This history goes back roughly 30 years if I recall correctly.

Mr. Hawkins fits into that mythical lineage of people who come from that most mysterious and phenomenal (and understated) places – Bell Labs. If you ask most people what the heck that place is or was, they would not know, despite what originated there.

Which is just about everything having to do with 20th century technology in one way or the other…. or close to it.

No, he didn’t work there, of course. I’m just saying it in a metaphysical sense.

His first job was at Intel – where they turned him down when he wanted to study brain science and he ultimately left to… well.. you can read about that.

NOTE: from hereon in, I’m going to use the same linguistic shortcut that Mr. Hawkins does when he uses the word BRAIN.

BRAIN – for the purposes of this discussion, means the cortex and not the entire human brain, which is not what Mr. Hawkins is trying (or has succeeded) to translate into intelligent machines.  This was his goal and he has succeeded thus far as Numenta proves.

As I understand from his book, neuroscientists and AI people were not even looking in the same place that Mr. Hawkins did (does) when it came to (comes to) understanding intelligence, the brain, and certainly not machine intelligence. 

In fact, MIT rejected him when he applied to work/study/teach there in the AI department because they just didn’t agree with his conceptualization of what machines could become.

He was correct and if I understand it correctly, they were wrong.

I have only skimmed the book so far since I got it last night,  I am SO completely enthralled by his vision and dedication, and unwillingness to be deterred by detractors who said that only AI  (traditional) could or would be the future of intelligent machines.

Mr. Hawkins rejected this notion AGES ago.

Just so I’m being clear, Mr. Hawkins is not talking about building (ever) anything that is a replicant of a human – he’s focused on the cortex – that area that is the seat of intelligence and learning. 

No Androids dreaming of electric sheep. Not even our beloved Mr. Data, I’m sorry to say.

Nothing that could pass the Turing test (if I’m correct) – which is a fascinating test but is an AI benchmark and not what Mr. Hawkins is striving for – or what he predicts intelligent machines will be used for or function as.

He’s very explicit about the aspects of ‘human nature’ that he’s leaving out – because he is focused on the cortex and the cortex is only a part of what makes us humans, or how human brains interact with our bodies, if you get my point.

It is far, far better if you let him explain why this was the case and still is the case – why AI will NEVER ‘complete its mission,’ as it were.

No amount of memory and processing power can make it so (yes, Jean Luc) because the manner in which AI folks (at MIT or wherever) were going about it missed the entire point of HOW the brain works.

Mr. Hawkins figured out how the brain works, and how it CAN be ‘translated’ into machine language (and I don’t mean that term in the IT definition of the word).

He didn’t START from the machine and work backwards and integrate, or try to make a machine behave like a human brain.

He worked the OTHER way around – first doing groundbreaking work in how the BRAIN works – as a hierarchical, memory-prediction system (he sticks to the cortex – the seat of ‘intelligence’ – which is an important distinction). 

Then, he figured out (and already obviously had vast IT experience) how to integrate that brain memory structure/design into ‘machines’  – he didn’t even have (or really need I suppose) an ‘actual’ machine at the time with which to test his (then) theory – he knew what he needed and how it would work and the industry would ‘catch up’ to him because of the inherent properties of ‘change over time’ that we all know exists in technology – ie – Moore’s Law, etc.

Well, MB turned into GB which turned into TB and more – we  ALL know that of course. I can carry a TB external HD in my back pocket (and even sit on it since it’s mil drop spec 🙂

So – the resources that he could not find to back him in the bad old days when he worked at Intel or even in the mid-1990’s and so forth – those hardware resources are now available, and at a fraction of the cost of what they used to be.

Now – there is GROK.

And GROK is ‘just’ a ‘phase one’ implementation of Mr. Hawkins’ ideas – but GROK has real teeth and you can use it and buy it and it works.

I cannot go into depth about how and why.  I am simply in awe of this visionary.

He stated (in On Intelligence) that parts of this theory (like all theories) could/might be proven inaccurate and need amendment or change.

But so it is with any scientific theory and the scientist who creates it.

He gives credit to the shoulders he stands/stood on, while supporting his main idea to the point of asserting that it WILL in fact revolutionize human understanding of how brains work, intelligence works, and ultimately what can and cannot be done (and is now being done ) in the new world of machine intelligence – Intelligent Machines (NOT, as he explicitly states, Artificial Intelligence or AI).

I don’t know if Numenta, Inc. will be the name of the company you see at the IPO 🙂 but if I were you,  I’d be on the lookout for whatever it does become.

If you DO keep an eye (through the newsletter, Twitter feed, Facebook, YouTube, etc.) on Mr. Hawkins, and this company and what he creates, you will most likely be watching the birth of a new era.

I don’t know how else to put it.

Here is one YouTube video to get you started. It’s certainly not bedtime reading, as it were. But it’s done well enough that even though the audience is a bunch of tech weenies (:-) I think that if you pay attention (and maybe know what binary is) you can follow along.

If not, then I will be posting other videos and sources of information about Numenta and his brain science theories and so forth. And I strongly recommend reading ‘On Intelligence,’ even though I would have loved to have edited it a bit since it is a bit awkward in terms of sentence construction. It’s probably intentionally written as such – since there was a NYT science writer editing it with him – but I’m still a bit surprised. He’s a fellow NY’er – and I’m sure it was meant to reflect his natural speech style. It’s not a big deal.

My memory-prediction system was predicting different linguistic patterns – that’s all it is. Otherwise, the ideas are Earth-shattering!

 

There is a TedTalks vid that I want to watch and probably will post as well.

I do hope that Mr. Hawkins will come to speak in NY sometime soon – I know this isn’t ‘new’ news- because analysts have been following him and his companies for ages now – but this IS a new iteration of his creativity. And it is the first beta-testing and testing of his GROK-ster.

GROKster the Rockster.

Advertisements
Leave a Comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: