THE RANDOMIZED SMOOTHING

Certified robustness that scales — not by bounding a network's Lipschitz constant, but by drowning it in Gaussian noise and counting votes. If the majority class survives noise of std σ with probability pA, the smoothed classifier carries a hard, closed-form L2 certificate: no perturbation smaller than R = σ·Φ-1(pA) can change its answer. Noise, turned into a proof.

SOURCE Cohen, Rosenfeld & Kolter, Certified Adversarial Robustness via Randomized Smoothing (2019) — arxiv.org/abs/1902.02918. Rendered, not quoted.

BLUE TEAM · BUILDS & DEFENDS
3
THE MODEL

A base classifier f is smoothed into g: to classify x, sample x + ε, ε ~ N(0, σ²I), and return the majority vote. Let pA be the probability the top class wins under that noise.

Theorem. g is constant inside the L2 ball of radius R = σ·Φ-1(pA) around x. No attack of L2 norm < R flips it.

Here Φ-1 is the inverse standard-normal CDF (a probit). The bound is tight: an adversary that only knows pA can force a flip at exactly R.

Ground truth is a constructed linear base f(x)=sign(w·x), w=[3,4], evaluated at x=[2,1]. Every quantity below is closed-form.

5
THE LINEAGE

Neighbour: the-certified-robustness — the Lipschitz-bound school (IBP, CROWN, Reluplex) that proves a no-attack zone by bounding how fast the network can change.

Smoothing pays a different price. It needs no Lipschitz constant and no access to weights — it works on models far too big to bound, at the cost of noise-time inference and a sampling estimate of pA. A probabilistic no-attack zone where a deterministic one was impossible.

7
THE WITNESS

Live re-check of the load-bearing identity R = σ·Φ-1(pA) against the linear model's exact margin, probed at a noise level σ=2 ≠ 1 so a dropped factor cannot hide.

WITNESS: verifying…

Green only while the certificate scales with σ. Trip THE TAMPER (window 6) and this flips red on the spot.

THE MACHINE
4
IN ↓DATA IN

The noise level and the measured top-class probability:

1.00
0.9772

Preset from the linear base at σ=1: pA=Φ(w·x / (σ ∥w∥)) = Φ(2) ≈ 0.9772.

smooth & certify
0
LITTHE PANEL · THE ENGINE
2.0000
certified L2 radius  R = σ·Φ-1(pA)
Φ-1(pA)  =2.0000
exact L2 margin of the linear base  =2.0000
identity R − margin  =0.0e+0

For the linear base the certificate is exact: smoothing recovers the true distance to the decision boundary. pA→1 ⇒ R→∞; pA=0.5 ⇒ R=0 (a coin flip certifies nothing).

···
proven
8
OUT ↓DATA OUT

At σ=1, pA=Φ(2) the smoothed classifier is certifiably unflippable inside R=2.0000 in L2 — and this equals the linear base's exact margin, so the bound is met with no slack. Raise σ and R grows for the same pA: noise buys certified radius, trading clean accuracy for it.

RED TEAM · ATTACKS & BREAKS
1
WALLTHE ADVERSARY
The certificate is a wall only in L2. Against the linear base, the closed-form optimal L2 attack is δ = −margin·w/∥w∥ = [−1.2,−1.6], norm exactly 2.0 — it reaches the boundary at precisely R, never sooner. But switch norms: the optimal L∞ attack ε·sign(w) flips the base at ε = margin / ∥w∥1 = 10/7 ≈ 1.43, whose L2 norm is only 1.43√2 ≈ 2.02 — barely outside, and an L∞ certificate here does not follow from the L2 one.

Gaussian smoothing certifies L2 and L2 alone. Other threat models need other noise (uniform / Laplacian) — and weaker guarantees.

2
THE GRAVEYARD

“A high certified accuracy means the model is actually that robust.”

→ It means robust with confidence 1−α from a Monte-Carlo estimate of pA. The certificate is probabilistic; a lower-confidence-bound on pA must be used or the radius over-claims. AMBER

“Bigger σ is strictly better — free robustness.”

→ Bigger σ enlarges R for a fixed pA, but it also drives pA down (the base sees noisier inputs), collapsing clean accuracy. The trade is real.

“Φ-1(1) is a big number.”

→ It is +∞. A never-wrong class is certified everywhere — which is exactly why the estimate never returns pA=1 from finite samples.

6
THE TAMPER

Disclosed planted void: drop the σ factor and report R = Φ-1(pA). It looks fine at σ=1 and over-claims for small σ. THE WITNESS (7) probes at σ=2 and catches it.