Indexing the archive…
Your Universe of Digital Possibilities
Store a few patterns in a net of neurons by one rule — wire together the cells that agree — then hand it a corrupted version of one. No index, no lookup: every neuron just keeps matching the majority around it, and the whole state rolls downhill in energy until it lands in the nearest stored pattern, cleaned. That is content-addressable memory — recall by resemblance, the cue is the address — and it is the closest the rack comes to a hard question: how does a network remember, and how is a memory — maybe a self — just a stable basin a tangle of connections falls into?
The Ising energy of The Threshold, but with learned couplings wij. Every neuron flip can only lower it, so the state slides downhill and must stop — in a memory.
Carve each pattern ξμ into the wiring: neurons that agree in a memory get a positive link, those that disagree a negative one. Memory is a property of the connections, never one address.
Show a corrupted cue, then let each neuron align with the majority vote of the rest. The state rolls down E into the nearest stored pattern — content-addressable memory: the cue is the address.
Store more than ≈0.138N patterns into N neurons and the basins overlap and collapse into spurious blends — the net forgets everything at once. The spin-glass price of a distributed memory.
A Hopfield net is an Ising magnet (INST·04) with the couplings learned instead of fixed: set J → wij and h → 0 and the energies are identical. Memory and magnetism are one piece of physics.
This is the rack’s memory instrument. The energy is the Ising Hamiltonian of The Threshold (INST·04) with learned couplings; recall is a walk down a landscape of minima, exactly the descent of The Descent (INST·27) — but onto a stored state, not a decision boundary. Add temperature and it becomes the Boltzmann sampler of The Anneal (INST·34), which can melt out of the spurious traps. Hopfield & Hinton took the 2024 Nobel in Physics for exactly this bridge — the physics of spins, run as the mathematics of memory.