Alphabeta Math
PropositionStatement: AI-generatedProof: AI-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Energy dissipation along a downward gradient curve

Statement

Let (M,g) be a smooth Riemannian manifold, let U:M→R be smooth, and let q:[0,∞)→M be a C1 downward gradient curve: q′(t)=−grad⁡gU(q(t)). At t=0 the derivative is understood from the right. The gradient is the vector field characterized by gx(grad⁡gU(x),v)=dUx(v) for every v∈TxM; write ∥v∥g2=gx(v,v). Then ddtU(q(t))=−∥grad⁡gU(q(t))∥g2, with a right derivative at 0, and U(q(t)) is nonincreasing. If U is bounded below along the curve, there is a finite limit L=lim⁡t→∞U(q(t)) and ∫0∞∥grad⁡gU(q(t))∥g2 dt=U(q(0))−L. Here the integral is the limit of the finite-interval integrals. Square-integrability of the speed is not asserted to imply finite length or convergence to a single point.

Facts & Assumptions

Given: The manifold, metric, potential and curve in the statement.

[F1]

The gradient is the metric dual of the differential, and the metric is positive definite (The Riemannian gradient is the metric dual of the differential, Riemannian metric and riemannian manifold).

[F2]

In Euclidean coordinates the derivative of a composition is the composition of the derivatives (The chain rule for total derivatives: D(g∘ f)(a)=Dg(f(a))∘ Df(a)).

[F3]

The fundamental theorem integrates the derivative of a continuously differentiable real function on an interval with endpoints in its domain (The fundamental theorem of calculus and integration by parts).

[F4]

A continuous real function on a closed interval, differentiable in its interior, has an interior derivative equal to its secant slope (Mean value theorem on a closed real interval).

[F5]

Every nonempty real set bounded below has a greatest lower bound (Every nonempty set bounded below has an infimum), and values of the set lie arbitrarily close above that bound (Epsilon characterisation of the infimum).

[F6]

On each finite restriction of q, use chartwise C1 regularity with continuous first derivatives and the one-sided endpoint convention; the derivatives match at interior subdivision points, so there are no interior corners (Piecewise c one curve on a manifold).

[F7]

In a smooth chart x=(x1,…,xn) around p, the coordinate derivation ∂i∣p acts on a smooth local function h by ∂i∣p([h])=∂i(h∘x−1)(x(p)), and these derivations form a basis of TpM (Coordinate derivations at a point, Coordinate derivations form a basis of the tangent space).

[F8]

The differential of a smooth real-valued function acts on a tangent derivation by dhp(v)=v([h]) (The differential of a smooth real-valued function).

Proof

Given: The manifold, metric, potential and curve in the statement.

Proof technique: direct.

Goal: Prove the energy dissipation identity, monotonicity and the stated limiting integral.

1.1givenF2F6F7F8constructalgebra

Fix t≥0, put p=q(t), choose a smooth chart x around p, and write y=x∘q near t. By [F6], y is C1, with the right derivative at 0. By [F7], Vt=∑iyi′(t)∂i∣p is a tangent vector. For any smooth local function h near p, [F7], [F8] and the Euclidean chain rule give dhp(Vt)=∑iyi′(t)∂i(h∘x−1)(y(t))=(h∘q)′(t) when t>0. At zero, substitute y(s)=y(0)+sy+′(0)+o(s) as s↓0 into the first-order expansion of h∘x−1 to obtain the same identity with right derivatives. These identities determine the action of Vt on every smooth germ by [F8], independently of the chart. Thus Vt is the intrinsic tangent velocity denoted by q′(t), including its right endpoint value.

2.1givenF1F2F6F8step 1.1algebra

Put f(t)=U(q(t)). Step 1.1 with h=U gives f′(t)=dUq(t)(q′(t)), with the right derivative at zero. The chartwise C1 curve and the smooth potential make this derivative continuous, including from the right at zero. Substituting the downward gradient equation and the metric-dual identity yields f′(t)=−gq(t)(grad⁡gU,grad⁡gU)=−∥grad⁡gU(q(t))∥g2≤0.

3.1step 2.1F4

For 0≤a<b, the mean value theorem applied to f on [a,b] gives f(b)−f(a)=f′(c)(b−a)≤0 for some c∈(a,b). Thus f is nonincreasing.

3.2step 2.1F3constructalgebra

For T>0, extend f to a continuously differentiable real function on an open interval containing [0,T] by setting f(t)=f(0)+tf+′(0) for t<0 near zero. The fundamental theorem and step 2.1 give ∫0T∥grad⁡gU(q(t))∥g2 dt=f(0)−f(T).

4.1givenF5step 3.1step 3.2algebra∎

Suppose f is bounded below and set L=inf⁡{f(t):t≥0}. This is finite because the set is nonempty and bounded below. For every ε>0 there is s≥0 with f(s)<L+ε; for all t≥s, monotonicity gives L≤f(t)≤f(s)<L+ε. Hence f(t)→L as t→∞. Taking the limit in step 3.2 proves the stated improper integral identity. Since q′=−grad⁡gU, its squared speed is the integrand.

Depends on

Used by · 3

Dependency tree · 2 levels

Each result is drawn at its shortest distance from this item. Arrows run from prerequisites to the items that use them; this item has a heavier outline. External source nodes link directly to the cited text. The view shows up to two dependency steps; open a prerequisite to continue.