Human Uniqueness or Simply Repeating Patterns ?

I will question the doctrine of the human uniqueness in order to find basic repeating patterns which are useful for AI-development in presentation. To do so will not be a huge issue. A lot of research is available that shows us what we need to know for understanding the way a Machine Human Level Intelligence could look like.

Afterwards we can discuss it. May will find something very clever :) I'm looking forward to meet you.

I've booked the Sherard-Hall for our meetup:

http://www.forumatgreenwich.org/roomshalls-for-hire/sherard-hall/

Here you can find all contact details:

http://www.forumatgreenwich.org/contact-us/

Join or login to comment.

  • James K.

    I was thinking that we were a bit limited for what we can do on meetup as a social network and forum.

    I suggest we think about creating a private group on linked in.

    We can then have a thread based discussion forum.

    Ideally I would like to see a tool developed similar to linked in that was more tailored towards developing group working, cooperation and information sharing in an academic area of study. The kind of thing we could then look to develop would be an open standard UML component model for an AI architecture made of pluggable components.

    Other useful information on a more developed site would be profiles on people, companies institutions projects and topics.

    Does such a thing already exists?

    I have a website where we could have a subdomain setup if someone knew of a better tool.

    Thoughts?

    (Hopefully fewer typos and mixed letters this time - yes I am dyslexic)

    March 4, 2013

    • Charles G.

      I'm not sure about an academic focused site, but there are plenty of software project sites with great collaboration tools; wiki's, forums etc.

      March 16, 2013

    • Charles G.

      One I've used before is https://www.assembla.c...­

      March 16, 2013

  • Charles G.

    If you haven't seen it there's an interesting google tech talk by Jeff Hawkins just out. http://www.youtube.com/watch?v=4y43qwS8fl4

    1 · March 11, 2013

    • Robert M W.

      It is indeed a very good talk from Jeff Hawkins and furthermore a good introduction into his approach. His prediction system is very intelligent and very robust. However, I think it could be even more improved by using different kind of entities. The personality types, we spoke about, could be used as a basic idea to develop the different kind of entities. Why could they be useful? People’s language and behaviour is based on their own personality type. It is shown the way they speak, dress or just behave. An artificial system which is internally structured according to these types should meet the requirements for HLMI more likely. But this is could already be the topic of a further meetup.
      Cheers
      Rob

      March 12, 2013

  • A former member
    A former member

    Might I suggest adjusting the scope of our discussion?

    The philosophical discussions suggested [on free will and on intelligence ] have high probabillities of failure. By which i mean be incapable of reaching a comprehension of intelligence; the best we might hope for being a restriction of our vocabulary upon which to further investigate. Such discussion only proves that whatever else it is, Artificial Intelligence is not a thing comparable to cognition.

    Equally, the point might not be to approach a comprehension of intelligence but an appreciation of our computational capabilities as programmers. Since many AI proponents have no predisposition toward humanity simulators but to the enabled analytical possibilities vast data manipulators and automation offers.

    I am likely to attend any and all of these again, and would be happy to aid in preparation of discussion materials, either way. It will be good to keep an open mind with our meetup.

    March 3, 2013

    • James K.

      I like this suggestion. To summarise what is practical and what is possible? To go further can we:

      Define "Packages" "Domains" "Sub-systems" or whatever you choose to call them to build a system.

      Discuss or even develop use cases and interfaces between these sub-systems.

      By doing this we could then perhaps focus on understanding specifics.

      1 · March 3, 2013

    • Robert M W.

      I agree to James and that's the intention of the meetup anyway like yesterday already out spoken.

      March 3, 2013

  • James K.

    On artificial conciousness:

    http://en.wikipedia.org/wiki/Global_Workspace_Theory
    http://en.wikipedia.org/wiki/Cognitive_architecture
    http://en.wikipedia.org/wiki/LIDA_(cognitive_architecture)

    Much more is available particularly at Memphis Uni a tutorial and java LIDA download are available.

    March 3, 2013

    • James K.

      I mentioned it during discussion about being probably the most successful computer model of artificial conciousness (links you quoted are what I meant).

      March 3, 2013

  • Tomi M.

    We could devote at least one meetup for discussing free will. I get very depressed if I think all our decisions have been predetermined even before we were born.
    Perhaps I'm waking an old monster discussion, but it seems there is an opportunity in nature to truly have free will. If this is the case with human or animal brains, the implication is that classical computer simulations miss something essential no matter how large they are. http://en.wikipedia.org/wiki/Free_will_theorem
    http://en.wikipedia.org/wiki/The_Emperor%27s_New_Mind

    March 3, 2013

  • A former member
    A former member

    Robot carers in the NHS and Medicaid are a partial answer to their management and financial problems. Especially with so many 'unwanted' old people in Japan and the developed West.

    I think they need to be able to understand spoken language better than they can at present. It cannot be so very very difficult can it?
    Computers have huge memory banks these days to hold simulated outcomes and even most of the common questions a patient would ask.

    I would like to contribute to such a project and be a member of a think tank to supply answers.

    March 3, 2013

    • James K.

      Absolutely right I think. But in my case it will be what my PhD will be on. So I cannot help much unless I fail to get an industrial partner (KTP scheme http://www.ktponline....­). Keep me informed though.

      March 3, 2013

  • Meneghin M.

    Hi everyone, many of you are very knowledgeable on this topic, sorry if I say something incorrect.

    The interesting meeting of today seemed to have this implicit message:
    At the moment we cannot have an AI equivalent to the human brain, but we can try to create one that appears similar to us. Assigning a personality trait to an AI seems to be a great achievement of it.

    If this is the goal the centre of interest moves to practical applications. In this case it could be interesting to see what some companies are doing. It's different, if the long term goal is to have AI similar to humans.
    For example emotions are essentials forces that drive our motivation for everything we do.
    Without this we may have AI very good for some specific tasks, but in a very narrow domain and without creativity.
    It may be very interesting to see if the neurosciences and the cognitive studies can give a little hint on how to model internal details of the human brain.

    Regards, Mauro

    March 3, 2013

  • Charles G.

    Thanks for the links to the books. People were asking about large scale neural simulations, a good example was in the press recently; http://nengo.ca/

    There's also a great talk on deep learning, from google which is worth watching if your interested in training neural nets. http://www.youtube.com/watch?v=VdIURAu1-aU

    March 2, 2013

  • James K.

    I believe the Jeff Hawkins book Robert referred to was call On Intelligence and can be found at: http://www.amazon.co.uk/On-Intelligence-Jeff-Hawkins/dp/0805078533/ref=sr_1_2?ie=UTF8&qid=1362260369&sr=8-2

    -- Which is excellent and I found fun to read.

    A more general and academic introduction to AI can be found in Pentti O'Haikonen's (Principle Scientist Cognitive Technology Nokia Research) book:

    http://www.amazon.co.uk/The-Cognitive-Approach-Conscious-Machines/dp/0907845428/ref=sr_1_sc_1?s=books&ie=UTF8&qid=1362260550&sr=1-1-spell

    This book covers lots and lots of ground. It struck me that it was a very good statement of requirements for true AI system. It is a much harder read than Jeff's book. It was O'Haikonen's book whcih I felt could be the basis on which an object orientated software design could be developed. This book has praise on the back cover from university researchers at Kings, Imperial and Birmingham.

    March 2, 2013

  • Phil S.

    Regret, now unable to attend.

    March 1, 2013

  • James K.

    If anyone wishes to search me out afterwards and discuss some research ideas please do. I have the offer of a PhD position working on the automated conversion of text descriptions to process simulations. My ultimate interest is in the development of a cloud based shared memory for artificial intelligent agents. I am looking for an industrial partner. (Apologies for the techie talk and the approach).

    February 23, 2013

  • Robert M W.

    here is a link with a short description how to get there:
    http://www.forumatgreenwich.org/how-to-get-here/ Hope that helps

    looking forward to meet you :)

    February 11, 2013

  • Robert M W.

    Hi Duncan, yes looking forward to meet you and to announce the venue. Unfortunately the Google campus is booked out for March 2nd but I'm optimistic to have found a nice venue by the beginning of next week. Cheers Rob

    February 10, 2013

  • Duncan C.

    I look forward to meeting everyone and hope to have an exciting time exploring the infinite possibilities.

    February 7, 2013

29 went

Our Sponsors

  • Telrock

    We have great meeting room venue in the heart of Shoreditch.

People in this
Meetup are also in:

Imagine having a community behind you

Get started Learn more
Henry

I decided to start Reno Motorcycle Riders Group because I wanted to be part of a group of people who enjoyed my passion... I was excited and nervous. Our group has grown by leaps and bounds. I never thought it would be this big.

Henry, started Reno Motorcycle Riders

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy