top of page
Search

"Recombinant dynamical systems" pre-print posted

  • May 20, 2025
  • 1 min read

This is an idea and a small but curious result that I've been sitting on for a very long time. Finally got around to sharing it. It the simplest possible model I could come up with for experience-based problem solving and how it might be implemented by a connectionist dynamical system. - Saul



Recombinant dynamical systems

We describe a connectionist model that attempts to capture a notion of experience-based problem solving or task learning, whereby solutions to newly encountered problems are composed from remembered solutions to prior problems. We apply this model to the computational problem of efficient sequence generation, a problem for which there is no obvious gradient descent procedure, and for which not all posable problem instances are solvable. Empirical tests show promising evidence of utility.

 
 

Recent Posts

See All
SurfDist paper published in WACV

Jackson Borchardt's paper "SurfDist: Interpretable Three-Dimensional Instance Segmentation Using Curved Surface Patches" was published in WACV 2026. It's an approach for segmentation of smooth blobs w

 
 

© 2016 onward   Foundations of Cognition Lab

bottom of page