Hi all, Born in 1970, I was a child surrounded by electro-mechanical gadgets that into the '80s were being miniaturized and consequently harder to take apart and understand how they work. In the mid-80's my first home computer, a Z80A 3.5 MHz 48KB beast (ZX Spectrum, competing with Commodore, Amiga, etc before Nintendo and Sega) led to lots of cassette tape swapping and game playing, plus an insight into the use of computers to aid artists and musicians. Early computer classes in secondary school involved writing out programs on grid paper to be sent to a local University to run on a PDP/11, a week later a print out of the program running would return. Subsequently a computer science BSc degree from a London, UK uni. concentrating on the study of global illumination simulation (radiosity within diffuse interactions, bi-directional reflection functions, etc.).
At the tail end of the early 90's economic recession I managed to get a job working on custom PC graphics cards doing low level coding (down to assembler and micro-coding, to hardware bootstrapping and debugging). A cheaper alternative for training simulators than SGI based machines, and one of the first 3D texturing capable PC cards with up to 16 boards connected in an industrial chassis attached to one display. A few years later I jumped industries to spend 14 or so years making computer and console games. The latter six as a contractor through a co-owned company acting as secretary, CTO & CFO Where we mainly provided bespoke art teams to temporarily expand game productions (Hollywood model), as well as assisting production of TV adverts, CGI for short film & music videos. After the demise and divorce of the company a few colleagues went on to successful feature film productions. Through three generations of consoles, from first games on PlayStation 1, PC (DirectX 3 then 5) and Nintendo GB and GBC, to Harry Potter, Sports, and the Wolfenstein franchise, the highlights are numerous. Always an engine coder, but also dabbling in audio and UI work. Mainly through C++ A year ago I read "On Intelligence" and it filled a void in my machine learning knowledge. A piece to a puzzle that had formed from endless disappointing stops off into AI research over the past few decades. Hoping to add behavior to my robotics projects. This last year has seen my research focus shift from visual cortical processing to mammalian auditory scene analysis. With an initial hope to instigate inter-aural time and phase differences within hierarchical temporal memory using cortical learning algorithms, and possibly later, cortical involvement in resonance. Real life has been throwing around some curve balls of late, but I'm hoping that this year life settles and more advances can be discovered. Best regards, Richard. https://github.com/rcrowder On Tue, Apr 14, 2015 at 7:26 PM, Matthew Taylor <[email protected]> wrote: > David Wood switched subjects about 2 hours ago to "Chess (was Re: > What's your technical history?)". You should have emails in your inbox > with this subject now, so you can just respond to them. > > --------- > Matt Taylor > OS Community Flag-Bearer > Numenta > > > On Tue, Apr 14, 2015 at 11:24 AM, cogmission (David Ray) > <[email protected]> wrote: > > Matt, > > > > How do I continue the conversation though? (Chess convo). That link > takes me > > to the list archive, but I want to mail to it... How do I do that? > > > > On Tue, Apr 14, 2015 at 12:56 PM, Matthew Taylor <[email protected]> > wrote: > >> > >> Hey folks, chess discussion has moved to > >> > >> > http://lists.numenta.org/pipermail/nupic_lists.numenta.org/2015-April/010770.html > >> --------- > >> Matt Taylor > >> OS Community Flag-Bearer > >> Numenta > >> > >> > >> On Tue, Apr 14, 2015 at 10:54 AM, cogmission (David Ray) > >> <[email protected]> wrote: > >> > @Ralf @Sergey @Matt > >> > > >> >> NuPIC is too slow to learn to play chess simply by remembering the > >> >> position of every piece - games are too different. > >> > > >> > > >> > I think the question was not how to teach the HTM to play chess, but > to > >> > predict the next moves of known openings? "Playing" is the next step, > >> > but > >> > not required to get something that is useful... no? > >> > > >> > > >> > -- > >> > With kind regards, > >> > > >> > David Ray > >> > Java Solutions Architect > >> > > >> > cortical.io > >> > Sponsor of: HTM.java > >> > > >> > [email protected] > >> > http://cortical.io > >> > > > > > > > > -- > > With kind regards, > > > > David Ray > > Java Solutions Architect > > > > cortical.io > > Sponsor of: HTM.java > > > > [email protected] > > http://cortical.io > >
