Well, you'll be *really *unimpressed with me, because ... I have no plan!

I'm kidding ... partly. I sit down at the beginning of every year and sketch 
out a development schedule for the next year. Then sometimes, after I actually 
start working, I end up changing my mind. And I definitely don't have 
everything planned out all the way to the completion of the project. Maybe my 
death (at least four decades hence, I hope) will be the completion of the 
project.

Working on Acuitas reminds me of pulling open a big composite flower ... like a 
dahlia.  Every time I peel a petal up, I can see three more underneath it.  
Instead of trying to surmise what the whole inside of the flower looks like, I 
just keep peeling and looking.  I get done with a mechanism, think about what I 
have, and then ask myself, "Okay, what does he need now?  Okay, how do I write 
that?"

Things on the docket for this year (unless I change my mind) include:
*More agent-like/goal-directed behaviors
*Some "theory of mind" stuff (modeling other agents)
*The beginnings of an ability to grasp narratives (Acuitas just processes 
single sentences right now)
*Text parser upgrades to handle conjunctions
*More ways to retrieve and use info from the episodic memory

Maybe I will draw a diagram of him someday, but I keep being too busy.
------------------------------------------
Artificial General Intelligence List: AGI
Permalink: 
https://agi.topicbox.com/groups/agi/Tb135fde0a846b51f-M9be8bddb36fb7a671825fafc
Delivery options: https://agi.topicbox.com/groups/agi/subscription

Reply via email to