Mercurial > cortex
view thesis/dylan-reject.diff @ 537:bc5eb476693a
improve confusing caption.
author | Robert McIntyre <rlm@mit.edu> |
---|---|
date | Sun, 27 Apr 2014 21:50:04 -0400 |
parents | ced955c3c84f |
children |
line wrap: on
line source
1 @@ -3234,8 +3354,8 @@3 - =CORTEX=, a system for creating simulated creatures with rich4 senses.5 - - =EMPATH=, a program for recognizing actions by imagining sensory6 - experience.7 + - =EMPATH=, a program for recognizing actions by aligning them with8 + personal sensory experiences.10 # An anatomical joke:11 # - Training13 @@ -2732,7 +2851,7 @@14 provided by an experience vector and reliably infering the rest of15 the senses.17 -** Empathy is the process of tracing though \Phi-space18 +** ``Empathy'' requires retracing steps though \Phi-space20 Here is the core of a basic empathy algorithm, starting with an21 experience vector:22 @@ -2440,7 +2558,8 @@23 hard control problems without worrying about physics or24 senses.26 -* Empathy in a simulated worm27 +* =EMPATH=: the simulated worm experiment28 +# Empathy in a simulated worm30 Here I develop a computational model of empathy, using =CORTEX= as a31 base. Empathy in this context is the ability to observe another33 @@ -436,7 +554,7 @@34 doing everything in software is far cheaper than building custom35 real-time hardware. All you need is a laptop and some patience.37 -** Because of Time, simulation is perferable to reality38 +** Simulated time enables rapid prototyping and complex scenes40 I envision =CORTEX= being used to support rapid prototyping and41 iteration of ideas. Even if I could put together a well constructed