XPost: sci.physics.relativity, comp.lang.prolog   
   From: janburse@fastmail.fm   
      
   Hi,   
      
   The bottom line is often, PRAMs might be   
   closer to physics. Especially for certain   
   machine learning algorithms or questions   
      
   from modelling perception or action. You   
   might get better results if you model the   
   problem in terms of Boltzman machines,   
      
   or whatever from the arsenal of physics.   
      
   Bye   
      
   P.S.: Whats was a little popular for a certain   
   moment of time, was also the idea of partical   
   swarm optimization, for machine learning or   
      
   for problem solving:   
      
   Particle swarm optimization   
   https://en.wikipedia.org/wiki/Particle_swarm_optimization   
      
   Not sure how much of it got supperseeded,   
   it mostlikey survides in AlphaEvolve by Google,   
   looks like a genetic algorithm thing, which   
      
   is another name for this "physics".   
      
   Maciej Woźniak schrieb:   
   > On 12/1/2025 5:12 PM, Mild Shock wrote:   
   >> Hi,   
   >>   
   >> Simulation is not so easy.   
   >   
   > I've never said it is easy. Some randomness   
   > or pseudorandomness existed for a long time,   
   > it's not enough for me to speak about a   
   > different architecture.   
   >   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)