Worked exercises in gauge structure

Insert ψₓ = gₓ⁻¹ψ′ₓ into the original relation and multiply the target by gᵧ:

Both endpoints matter. A loop based at x has the same initial and final frame, so the internal frame changes cancel and W′ = gₓWgₓ⁻¹. Cyclicity of the trace then proves Tr W′ = Tr W.

Continue with the related article or explorer →

Expand the endpoint law to first order:

Matching I − A′dx gives A′ = gAg⁻¹ − (dg)g⁻¹. This agrees with D′(gψ) = gDψ for D = d + A. A convention with a Hermitian physical potential or inverse frame coordinates must be converted before comparing signs.

Continue with the related article or explorer →

The identity Re Tr W/2 = 1 − 2sin²a sin²b gives −1, so W = −I for this family. The Wilson cost is 1 − (−1) = 2. The dark-port probability is [1 − (−1)]/2 = 1, while the forced-return probability is |−1|² = 1.

At W = I the dark port instead has probability zero but the forced-return probability remains one. Squaring the amplitude has discarded the sign that the interference reference can detect. Verify this with the central-loop and commuting presets.

Continue with the related article or explorer →

Use E² = E, E(id − E) = 0 and (id − E)² = id − E. The product is E + exp[−γ(t+s)](id − E), so the channels form the stated semigroup. At γt = log 100, the remaining coefficient is 1/100.

This is a quantitative approximation to the declared completion channel. A finite positive time does not make the residual coefficient exactly zero, and the channel is not a reversible unitary conjugation. The full source also specifies which carrier correlations it retains.

Continue with the related article or explorer →

The real dimension of U(n) is n². Thus the combined carrier gives 25 dimensions and separate blocks give 9 + 4 = 13. Fixing the product determinant imposes one real phase condition, leaving 12. Fixing each determinant separately removes two, leaving 8 + 3 = 11.

The difference is information retained, not an arithmetic inconsistency. A combined metric permits cross-block mixing. Separate blocks remove that mixing; volume tensors then remove one or two independent phase freedoms.

Continue with the related article or explorer →

The residue is q + 2t₃ + 3t₂ = −4 + 4 + 0 = 0. The generator of the kernel therefore acts trivially on this representation, so it descends. A conjugate triplet must use triality two rather than the fundamental triality one.

For Q with q = 1, t₃ = 1 and t₂ = 1, the sum is six and also vanishes modulo six. Changing only Q’s charge to zero gives residue five and fails this global test, even though the dimensions of the representation spaces have not changed.

Continue with the related article or explorer →

The secular sum is strictly decreasing for z > λ*. At the lower endpoint, a term corresponding to λ* already contributes one to (Δ/4m)Σ(z − λ)⁻¹, and the other terms are positive. At the upper endpoint every denominator is at least Δ, so the whole sum is at most one.

Continuity and monotonicity give a unique crossing in the interval, including any equality at the upper endpoint. This bracket selects the physical eigenvalue without choosing among ambiguous polynomial roots. At t = 0 all free eigenvalues vanish and z = Δ exactly.

Continue with the related article or explorer →

Choose X = diag(θ, −θ, 0). Then Tr X² = 2θ², so the convention ΔE = ½κ̂₃Tr X² + O(X⁴) gives ΔE = κ̂₃θ² + O(θ⁴). The finite-difference ratio ΔE/θ² therefore tends to κ̂₃, not twice that value.

At t/Δ = 0.1, the exact coefficient is approximately 0.00006277560191Δ. The leading weak-hopping coefficient is 2(0.1)⁴Δ/3 ≈ 0.00006666666667Δ. Their difference is expected at nonzero hopping because the latter keeps only the leading term.

Continue with the related article or explorer →

The first block contributes 3 × 4 = 12 and the second 2 × 9 = 18, so Tr Q² = 30. If the block quadratic form carries the respective response weights, its value on Q is 12κ₃ + 18κ₂.

For equal penalties and hopping in the exact model, κ̂ₘ is proportional to 1/m, so κ̂₂/κ̂₃ = 3/2. These statements concern a specified trace convention and model response. A measured coupling requires the additional action and scale matching.

Continue with the related article or explorer →

Differentiation gives (dP)P + P(dP) = dP. Multiply on both sides by P to obtain 2P(dP)P = P(dP)P, hence P(dP)P = 0. The connection on a section s = Ps is ∇s = P ds, which differentiates the moving section as well as its subspace.

The curvature P(dP ∧ dP)P contains two derivatives and can connect through the complementary space between them. A vanishing once-sandwiched derivative therefore does not imply zero curvature. In a local frame, A = u†du provides the corresponding potential.

Continue with the related article or explorer →

The continuous phase is −2πb = −π/2 ≈ −1.570796327. At N = 7 the forward product gives phase approximately −1.487567709 and survival probability 0.3445680152. At N = 16 the phase is approximately −1.555509589 and survival probability 0.6291609396.

The probability tends to one and the phase tends to the continuous value as the same loop is sampled more finely. The count N is a discretization input. It is independent of the interpretation of the six inner generative levels.

Continue with the related article or explorer →

The mixed coefficients also vanish: 2(1) − 4 + 2 = 0 for SU(3)²U(1), and 3(1) − 3 = 0 for SU(2)²U(1). There are three colored quark doublets and one lepton doublet, an even total of four for the stated Witten test.

The representations and chirality convention were supplied before doing the sums. Cancellation verifies these inputs rather than deriving them.

Continue with the related article or explorer →

The weak mixed coefficient becomes 3(1) − 2 = 1. The gravitational coefficient changes by 2[−2 − (−3)] = 2. The cubic Abelian coefficient changes by 2[(−2)³ − (−3)³] = 2(−8 + 27) = 38.

The SU(3) coefficients remain unchanged because the lepton doublet is a color singlet. The fixed non-Abelian representations still contain four weak doublets, but L’s quotient residue changes as well. Local anomaly sums and global quotient descent are separate diagnostics.

Continue with the related article or explorer →

Since sin x = x + O(x³), λ₀(θ) = θ²/L² + O(θ⁴/(N²L²)). Its value approaches zero with the twist, even when N and L are fixed. For the neutral sector θ = 0 it is exactly zero.

Compactness of the circle alone therefore does not give a lower bound uniform over all charged twists approaching identity. A restricted fiber window can have a bound under extra assumptions, but the physical Hamiltonian still requires vacuum and mixing control.

Continue with the related article or explorer →

The high block is positive, d − z = 4. The Schur complement is F(0) = 1 − 6.25/4 = −0.5625, so H is not positive semidefinite. Direct diagonalization gives λ₋ = (5 − √34)/2 ≈ −0.415475947.

Both diagonal blocks are positive, but the mixing defeats the low bound. This is the simplest illustration of why eliminating a high sector must keep the induced negative term and control it quantitatively.

Continue with the related article or explorer →

Here r − t = 0.6, so A = D = 0.3 and B = −0.3. Then A − B²/D = 0.3 − 0.09/0.3 = 0. The vanishing value expresses the projected vacuum’s zero-energy completion.

Using the uncorrected off-diagonal mixing gives (1 − t)²(1 − p)/(r − t) = 0.5333333333, which is larger than the correct 0.3 by the ratio 16/9. The punctured low window has rank zero in this example, so no positive nontrivial low-window bound follows from the puncture alone.

Continue with the related article or explorer →

Each of the finitely many terms tends to one, so the limit is M. A divergent power law s⁻ᵈᐟ² cannot arise from this fixed finite list. Such an asymptotic requires the corresponding infinite-dimensional trace, a growing cutoff with a stated limiting order, or another explicitly defined normalization such as a per-volume trace.

The order of limits is part of a spectral-action argument. A finite numerical fit over a chosen range is useful evidence for that range, but it is not by itself an infinite-spectrum theorem.

Continue with the related article or explorer →

For each finite n the excitation gap is 1/n > 0. The infimum over the family is zero, so no positive n-independent lower bound follows. A uniform continuum claim requires a different estimate or a physical rescaling justified by the model’s unit convention.

This simple sequence explains the logical role of uniformity without purporting to model the gauge dynamics. In the actual programme, volume, spatial spacing, auxiliary spacing, physical time step and retained sectors all have to be tracked in the required order.

Continue with the related article or explorer →

Both partial traces equal I₂/2. In the Bell basis the eigenvalues are (1+3s)/4 and three copies of (1−s)/4. Hence Tr ρ² = (1+3s²)/4 = 7/16, while the joint zz correlation is s = 1/2. The unchanged marginals therefore do not fix this joint statistic. They also do not prove that the channel fixes each full carrier algebra for arbitrary inputs.

The partial transpose has three eigenvalues (1+s)/4 and one eigenvalue (1−3s)/4. Its negativity is max[0,(3s−1)/4]. It vanishes once s ≤ 1/3, so the threshold is γt = log 3. This finite entanglement threshold differs from exact depolarization: s remains positive at every finite time when γ is finite and positive.

At the origin Fxy = diag(2iβ,−2iβ), so −Tr(Fxy²) = 8β². The two cases give invariant values 8 and 0. In both cases H has eigenvalues 0,0,2,2. Equal eigenvalues and equal separating gaps have not selected equal bundle curvature.

The diagonal connection integrates to diag(iπ,−iπ), since the flux magnitude is 2πR²/(1+R²) = π. With D=d+A, parallel transport is exp(−∮A), so W=diag(exp(−iπ),exp(iπ))=−I₂. The determinant holonomy is one and Tr F is zero, while the rank-two holonomy is nontrivial. This is separate from the determinant-line flat-holonomy counterexample on a circle.

Set a=r−t₁=0.6 and b=r−t₂=0.3. Then k=a cos²α+b sin²α=0.45, the high block is D=pk=0.225, and c=ab/k=0.4. The exact low Schur matrix is diag(0,0.4). Its zero direction is the projected vacuum; the punctured low space is spanned by the second coordinate and has rank one.

The gap is −log(max(0.2,0.5)) = log 2 ≈ 0.69314718. The threshold energy is −log(0.8) ≈ 0.22314355, leaving an energy margin log(1.6) ≈ 0.47000363. The positive Schur coefficient 0.4 is a dimensionless threshold-form quantity, not that logarithmic energy margin. Converting either into a physical prediction requires an independently specified time step and model correspondence.