◄ WORLD II · THE FOLDTHE OCHO · blue builds │ the machine │ red breaks

THE BOLTZMANN MACHINE

Take a Hopfield net and let it gamble. Each unit flips on or off with probability σ(ΔE/T) — a coin whose bias is set by the energy it would save and by a temperature T. Run the coins long enough and the states you visit are not arbitrary: their frequencies settle onto the Boltzmann distribution, p(s) = e−E(s)/T/Z. Cool toward T→0 and the whole cloud collapses onto the single lowest-energy state. Down the center, data flows: the net and temperature go in, the sampler runs, the settled distribution comes out. The blue team builds and defends it; the red team tries to break it.

source Ackley, Hinton & Sejnowski, A Learning Algorithm for Boltzmann Machines, Cognitive Science 9(1):147–169 (1985). No scan of the paper itself sits on archive.org; the nearest item is the contemporaneous RSRE memorandum presenting the algorithm — archive.org/details/DTIC_ADA160644 (1985). Citation marked AMBER. Rendered, not quoted.

◧ blue team · builds & defends
3

THE MODEL — energy, then a coin

Four binary units, symmetrically wired (wij=wji, no self-loop). A state s has an energy

E(s) = −( Σi<j wij si sj + Σi bi si )

To update unit i, read its local field zi = Σj wijsj + bi (this equals the energy saved by turning it on), then set it on with probability σ(zi/T). Sweep every unit, repeat. This Gibbs sweep has exactly one stationary distribution — the Boltzmann one. Nothing is looked up; the histogram is earned, one biased coin at a time.

This net: all wij=1, all bi=0.5 — so more units on is always lower energy, and the unique floor is all-on.

5

THE LINEAGE — from Hopfield AVAN

Set T→0 and σ becomes a hard step: each unit just moves to lower energy, deterministically. That is exactly Hopfield's net — a machine that only ever rolls downhill and freezes in the first minimum it meets.

The Boltzmann machine is that net with the temperature turned back up: the same energy landscape, but now the coin can climb out of a shallow trap. Anneal T downward and it settles into deep minima, not the nearest one. This one idea — a stochastic, temperature-controlled energy net — is the ancestor of the deep belief net and the whole family of energy-based models. Each sphere is the next one's premise.

7

THE WITNESS live

The blue team's live check: draw a fresh fixed-seed sample at a diagnostic T=2.5 and confirm the histogram matches the analytic Boltzmann law within tolerance — plus that cooling concentrates on state 1111. If red drops the 1/T, this badge is where it shows.

▼ the machine ▼
4

DATA IN — the net & the heat in ↓

What you feed the engine is fixed and small enough to enumerate exactly: four units, six symmetric weights, four biases, and one dial — temperature. The 16 possible states and their energies are ground truth the sampler must reproduce:

units on (k)01234
energy E0−0.5−2−4.5−8
# states14641

Only the count of on-units sets the energy here (all weights equal), so the analytic law is checkable by hand. The dial T is the only free input to the panel below.

▼   feed the net + T into the sampler   ▼
0

▣ THE PANEL — the sampler LIT

1.5

partition Z(T) =   = Σ16 e−E/T
lowest-energy state: 1111  ·  E = −8
sampler vs law, max gap: (40k sweeps, seed-fixed)
stateEp(s) lawsampled

green bar = sampled frequency · grey = Boltzmann law. Move T — every number is recomputed and re-sampled on the spot.

▼   the sampler emits a settled distribution   ▼
8

DATA OUT — the settled law out ↓

What the machine produces, proven: a histogram of visited states that tracks p(s)=e−E/T/Z to within a few parts per thousand at every T, and that collapses onto the single minimum-energy state 1111 as the dial cools toward zero. No frequency is baked; each is counted from live coin-flips.

The blue team's witness (left) re-confirms the fit live; the red team (right) tries to make the histogram lie.

red team · attacks & breaks ◨
1

THE ADVERSARY

WALL This 4-unit toy mixes in milliseconds — the real machine does not. Boltzmann learning needs the equilibrium distribution at every weight update, and on any interesting net that sampling mixes so slowly it was effectively impractical. That single failure is why the method was largely abandoned for a decade until the restricted Boltzmann machine and contrastive divergence (Hinton, 2002) sidestepped full equilibrium.

And convergence here is only in the limit: 40k sweeps gets ~0.004 max gap, not zero. Finite samples always leave a residue — honest, and shown as the live "max gap" number, never rounded to nothing.

2

THE GRAVEYARD

"Sampling gives the exact Boltzmann distribution." Cut. Only as sweeps→∞. Any real run returns an approximation — the panel prints the residual gap so the approximation is never hidden.

"Cooling always finds the global minimum." Cut. Only slow enough annealing does; drop T too fast and it freezes in a local trap. This net has a single minimum, so it is honest here — a general net is not.

"The Boltzmann machine learns quickly." Kept, corrected. The learning rule is elegantly simple (⟨sisjdata − ⟨sisjmodel) but was notoriously slow — two sampling phases per step.

6

THE TAMPER — break it

The red team's move: delete the 1/T in the update — sample as if the machine were always at T=1, whatever the dial says. The histogram still looks like a valid Boltzmann distribution, just the wrong one. The blue team's witness (window 7) is watching.

Drop the temperature from the exponent and the sampled cloud freezes at the T=1 shape — at any other dial setting it no longer matches p(s)=e−E/T/Z. The witness recomputes, disagrees, and turns red. Nothing is faked; the attack is real and it is caught.