Generator

filter-misfit

One function in the lsqrgmres library, called 8 times across 7 essays. Below: what it draws at its defaults, what it draws at every value an essay asks for, the 6 claims it put to the test while drawing them, and where it stands against the rule this site is named for.

At its defaults it draws the weights 8 steps apply, on an operator 8.6% asymmetric. Two sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.

filter-misfit is one function in lib/figures/lsqrgmres.js — the other two krylov regularisers — one sequence, two recurrences, and a weight that stops being a function of σ. Everything below came out of it during this build, at arguments taken from the essays rather than invented for this page. A figure here is the figure a reader meets in an essay, and if the generator changes, this page changes with it.

At its defaults

Drawn even though every essay passes arguments — which on this site is every essay, at 100% of placements since the standard pass. A default nothing exercises is a trap for the next essay to call this with none, and this is the page where a default that has drifted from the figures around it becomes visible.

The weights 8 steps apply, on an operator 8.6% asymmetricTwo sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.036912151821242700.250.50.7511.25index kweight appliedonebidiagonalArnoldiis it a function of σmisfit, even fit1.6·10⁻¹²misfit, general fit0.063‖A − Aᵀ‖/‖A‖0.086one method's weights do not notice the operatorand the other's stop being a function of σ

Two sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.

asym: 0

The arguments are the ones Rank is a decision passes. A value drawn at the generator's defaults instead would be a picture no essay asked for and no assertion has been run against.

The weights 8 steps apply, on an operator 0.0% asymmetricTwo sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 8.7·10⁻¹³. The Arnoldi method's lie on a general polynomial to 7.2·10⁻¹³ — which on a symmetric operator is the level of rounding and on this one is 0%.036912151821242700.250.50.7511.25index kweight appliedonebidiagonalArnoldiis it a function of σmisfit, even fit8.7·10⁻¹³misfit, general fit7.2·10⁻¹³‖A − Aᵀ‖/‖A‖0one method's weights do not notice the operatorand the other's stop being a function of σ

Two sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 8.7·10⁻¹³. The Arnoldi method's lie on a general polynomial to 7.2·10⁻¹³ — which on a symmetric operator is the level of rounding and on this one is 0%.

asym: 0.518

The arguments are the ones Symmetry is worth more than precision passes. A value drawn at the generator's defaults instead would be a picture no essay asked for and no assertion has been run against.

The weights 8 steps apply, on an operator 51.8% asymmetricTwo sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.5·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.36 — which on a symmetric operator is the level of rounding and on this one is 36%.036912151821242700.250.50.7511.251.51.75index kweight appliedonebidiagonalArnoldiis it a function of σmisfit, even fit1.5·10⁻¹²misfit, general fit0.36‖A − Aᵀ‖/‖A‖0.52one method's weights do not notice the operatorand the other's stop being a function of σ

Two sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.5·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.36 — which on a symmetric operator is the level of rounding and on this one is 36%.

m: 8

The arguments are the ones The basis decides what a filter is passes. A value drawn at the generator's defaults instead would be a picture no essay asked for and no assertion has been run against.

The weights 8 steps apply, on an operator 8.6% asymmetricTwo sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.036912151821242700.250.50.7511.25index kweight appliedonebidiagonalArnoldiis it a function of σmisfit, even fit1.6·10⁻¹²misfit, general fit0.063‖A − Aᵀ‖/‖A‖0.086one method's weights do not notice the operatorand the other's stop being a function of σ

Two sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.

asym: 0.086

The arguments are the ones The matrix that is one row passes. A value drawn at the generator's defaults instead would be a picture no essay asked for and no assertion has been run against.

The weights 8 steps apply, on an operator 8.6% asymmetricTwo sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.036912151821242700.250.50.7511.25index kweight appliedonebidiagonalArnoldiis it a function of σmisfit, even fit1.6·10⁻¹²misfit, general fit0.063‖A − Aᵀ‖/‖A‖0.086one method's weights do not notice the operatorand the other's stop being a function of σ

Two sets of points against the singular-value index, each with the best polynomial of its own parity drawn through it. The bidiagonal method's weights lie on an even polynomial in σ to 1.6·10⁻¹². The Arnoldi method's lie on a general polynomial to 0.063 — which on a symmetric operator is the level of rounding and on this one is 6%.

What it checked while drawing

Every figure above asserted its own claims on the way to being drawn, and a claim that failed would have failed the build rather than drawn a wrong picture. Those assertions used to leave no trace at all: a passing one returned true and the only evidence the figure had checked anything was that nothing crashed. The list below is what they actually said, collected by running this generator with an observer installed — not a description of what it is believed to check.

6 distinct claims across 5 sets of arguments, grouped below by shape — because most of them are one sentence with a different number in it, and how many separate times that sentence was put to the test is the informative part.

a size the reference SVD is affordable at

a step count both methods are still informative at

and on a normal operator the Arnoldi method's are a polynomial too

and the Arnoldi method's are not a function of σ once A is not normal

one of the three operators this family has

the bidiagonal method's weights are an even polynomial in σ

Against the rule

It draws a decomposition and prints its residual. It calls lstsqQR, and every figure above carries the badge — which residualcheck verifies by looking for it in the emitted SVG rather than by finding the call that builds one. A badge that is constructed and then left out of the body is the failure that check exists for.

Across the library: the rule bites on 70 of 151 generators — 55 print a residual and 15 are exempt with a published reason; 81 factorise nothing. Read from lib/residual-rule.js, which is the same body the gate enforces from, and the gate's last check fails the build if this page and it disagree about any generator.

Where it is called

Changing this generator changes every figure on this list. That is what makes the list worth publishing rather than keeping in a check script.

Eigenvalues, singular values, rank

Rank is a decision

A floating-point matrix does not have a rank. It has a spectrum of singular values, and somewhere in that spectrum is a place where the values stop being signal and start being noise. Deciding where is a judgement, and the evidence for it is a gap.

Eigenvalues, singular values, rank

Symmetry is worth more than precision

A symmetric matrix gives up its eigenvalues to full accuracy however ill-conditioned it is. An unsymmetric one can move them by the eighth root of a perturbation, so the rounding involved in merely storing the matrix shifts the spectrum by a hundredth.

Regularisation, and the answer that is chosen

The basis decides what a filter is

The vocabulary of regularisation is spectral — a method keeps a component or discards it, and the weights are a function of the singular value. Row-normalising a symmetric blur so that it preserves a constant makes it 8.6% asymmetric, and that is enough to move GMRES's weights from 7·10⁻¹⁴ off a function of σ to 4.4·10⁻².

Two errors, and whose fault they are

The condition number is an amplifier

κ is usually introduced as a definition and then quoted. It is a measurement: perturb the input by a known amount, look at how much the output moves, and the largest ratio you can find is the number.

Eigenvalues, singular values, rank

The form a real matrix can reach

A real matrix with complex eigenvalues has no real triangular form, and the reason is one line — a real triangular matrix has a real diagonal, and a similarity does not move the spectrum. What it has instead is triangular except for one two-by-two block per conjugate pair, and the count is decided by the matrix rather than by where the iteration stopped.

Structure, and the solver that cannot see it

The matrix that is one row

A circulant of size 16 is sixteen numbers, has no zero entry anywhere, and hands over its entire spectrum in closed form — the discrete Fourier transform of its first column, exactly. An eigensolver spends a sweep of Jacobi rotations over 256 entries arriving at the same answer, and agrees to 1.2·10⁻¹⁵.

Iterating, instead of factorising

The spectrum that predicts nothing

For a symmetric matrix the eigenvalues govern how fast an iteration converges. Drop symmetry and they stop governing anything — there is a matrix whose eigenvalues are as evenly spread as eigenvalues can be, on which GMRES makes no progress at all until the last possible step.

The whole library · All essays · What must fail