Hasty Briefsbeta

Is life a form of computation?

3 hours ago
  • #computation
  • #AI
  • #biology
  • John von Neumann established a deep link between life and computation, showing that reproduction can be carried out by machines following coded instructions, similar to DNA.
  • Biological computing is massively parallel, decentralized, and noisy, differing from traditional digital computing which relies on centralized, sequential execution of instructions.
  • DNA operates like a program, with processes such as epigenetics and gene proximity effects adding complexity beyond simple digital computation.
  • Randomness in biological computing is a feature, not a bug, and is also utilized in computer science algorithms, as seen in early computers designed by Turing.
  • Modern AI relies on parallelism and randomness, exemplified by stochastic gradient descent and GPU parallelism, aligning more closely with biological computing models.
  • Von Neumann and Turing explored non-centralized computing models, such as cellular automata and neural networks, which mimic biological systems.
  • Cellular automata and neural cellular automata (NCAs) demonstrate how complex, lifelike behavior can emerge from simple, local rules, offering insights into the computational nature of life.
  • The first successful emulation of von Neumann’s self-replicating automaton in 1994 highlighted the computational complexity and inefficiency of simulating parallel systems on serial computers.
  • NCAs, combining neural nets with cellular automata, can model cell behaviors and regeneration, showing how local computations can produce global, lifelike patterns.