Saturday, August 29, 2015

Wargaming and WWII

I heard Norman Friedman speak on wargaming and world war II.  He made the point that the US could hope to cut off Japans overseas resupply and, if that didn't end the war, burn them out with bombing.  He then argued that the japanese had no such endgame thought out.  But I think the japanese believed that the US would simply tire of a long war (which they did) and then sue for peace on terms favorable to japan (you know, like what happened with vietnam a bit later).  Perhaps its Friedman who is just not thinking like the enemy thinks.

Thursday, August 20, 2015

Braitenberg vehicles with an Asa H brain

Braitenberg sketched out what might be a possible history of the evolution of intelligence. (Vehicles, MIT Press, 1986)  Some fraction of these different vehicle types were then subsequently fabricated using the earliest versions of Lego mindstorms. (Hogg, Martin, and  Resnick, Braitenberg creatures ,1991) The order in which concepts are learned and capabilities developed is important.  Braitenberg's book suggests a bit about some possible development sequences.  I believe that I have now been able to implement essentially all of Braitenberg's creatures using Asa H as the controller of Lego NXT 2.0 hardware.

Wednesday, August 19, 2015

More on Asa's concept of itself

As Asa H builds up its model of the world and of itself (4 March 2015 and 28 April 2015 blogs) there is not always a clear distinction between the two.  For example, Asa has just learned the temporal sequence:

detect light source
turn toward light source
approach light source
light intensity increase
Asa temperature increase

Tuesday, August 18, 2015

Sense of smell

Asa H does not have a true sense of smell but I do have a smoke detector/alarm input.

Humidity

When you get off the plane in Orlando or Singapore the first thing that you notice is the humidity.  I am adding a Vernier humidity sensor to my AI Asa H so that it can experience a similar sensation.

Sunday, August 16, 2015

Internet of things

If an internet of things develops and becomes as ubiquitous as some people believe it might serve as a good I/O system for my AI Asa H.

Saturday, August 15, 2015

Telepathy

While I do not believe in human telepathy my Asa H agents can telepathically communicate with one another.  (Mostly via Wi Fi and over the internet.)

Sunday, August 9, 2015

Taste

I have added Vernier pH and salinity sensors in order to give Asa H a simple sense of taste.

Tuesday, August 4, 2015

Kinds of thinking

I'm confident that we have built and now have in operation true thinking machines.  I am not as sure that we have identified and implemented all the kinds of thinking that might be possible.  One of the places that I still go looking is amongst the logics:
spatial
temporal
defeasible
paraconsistent
deontic
illocutionary
quantum
multi-modal
epistemic
intuitionist
fuzzy
probabilistic
substructural
categorical
higher order
etc,

Saturday, August 1, 2015

Training Asa H in real and virtual worlds

After each training session we typically record Asa H's hierarchical memory ( see, for example, blogs of 10 Feb. 2011 and  26 Aug. 2013).  This can then be reloaded to begin a new training session if we so choose.  Some training can be in real worlds involving things like Lego NXT robots and sensors.  Other training can involve simulated robots in virtual worlds.  It is quite possible to alternate between real worlds and simulations (virtual learning environments) as the memory is grown but one must be careful to avoid such things as sensory mismatches between the two sorts of robots. Adjusting the time scale of each of the training sessions is also tricky.