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.
Weighted AM-GM inequality with rational weights
Statement
Let with , let with , and let be rational weights with (Order on the rationals) whose images under the canonical embedding (The unique embedding of ℚ into an ordered field), written with the same symbols in field arithmetic, satisfy . Then
where is the rational power of Rational powers of a positive base.
Both sums are sums in , and that is not a detail. This library defines only for a sequence (Finite sums and finite products, by recursion); there is no finite sum of rationals and none is used here. The weights are therefore summed after being carried into by , and no step below sums anything outside . Nothing is lost by this reading, because is an injective field homomorphism: for the field equality is equivalent by injectivity to the same equality in , and the conclusion reads . Throughout the statement and proof, a rational used as a field coefficient or a summand denotes its image under the bare inclusion (Finite sums and finite products, by recursion); a rational used as an exponent retains its original rational type. Natural list lengths and index bounds also retain their natural type.
Why the weights are rational. This theorem uses the rational powers of Rational powers of a positive base, which are the powers available at this point in the reading order. Arbitrary real weights are outside this theorem's scope; real powers for positive bases are defined later through the exponential and logarithm in the later definition of real powers. That later construction is the library's chosen development, rather than a logical necessity for defining real powers or proving their laws, as the closing remark of this page explains. Taking and recovers the two-term case of The arithmetic mean, geometric mean inequality.
Facts & Assumptions
Given: A natural , reals , and rationals with , a sum in .
AM-GM (The arithmetic mean, geometric mean inequality): for with , .
Finite sums and products, defined ONLY for sequences (Finite sums and finite products, by recursion, Laws of finite sums and finite products): the recursion clauses and ; splitting of sums and products at any index , , where the tail is by definition the shifted sum (Finite sums and finite products, by recursion), and likewise for products; scaling ; and the constant sum . Every and written below is therefore a sum or product in .
Constant product: , by induction on from the recursion clauses and , with both sides equal to at (Finite sums and finite products, by recursion, Integer powers , The principle of mathematical induction).
Rational power laws (Laws of rational exponents, Rational powers of a positive base, Existence and uniqueness of -th roots: a unique with ): for and rationals : , (hence, by induction on the number of factors, , using also the constant product for the empty case), and .
Rational arithmetic, the order on and the embedding (Every rational has a positive-denominator representative, The rationals as equivalence classes of pairs of integers, Arithmetic on the rationals, Order on the rationals, The unique embedding of ℚ into an ordered field, Canonical naturals are positive and strictly increasing): on a representative with positive denominator, holds exactly when in (Order on the rationals, with the order on as in Order on the integers); a nonnegative integer is the image of a unique natural and a positive integer the image of a unique natural (The naturals embed in the integers), which is the step that licenses reading such an , and such a denominator, as a natural number; every single rational has a representative with positive denominator (Every rational has a positive-denominator representative states exactly this, for one rational; the passage to a common denominator for finitely many is NOT quoted from it and is carried out by the induction inside step 1.1); is an injective order-preserving field homomorphism and therefore preserves zero, addition, products and inverses. Rational symbols in real sums and scalar products below denote their images through this map. On an integer the embedding is ; its image is positive for natural , and this map is injective on all of , including zero. Injectivity is used to pass from equality of field images to equality of natural numerator sums.
Induction principle (The principle of mathematical induction), used for the routine inductions on the number of terms below, and the recursion theorem (The recursion theorem), which is what defines a function on by a recursion clause.
Proof
Common denominator, by an induction written out rather than asserted. The claim at is: any rationals admit a natural and integers with for every . At take , there being no to produce. Assume the claim at and let be given: applying it to yields a natural and integers with , and has a representative with , hence with a natural ; then is a natural , and the integers for and satisfy and , since for . The claim at , applied to , fixes a natural and integers with . Finally each is a natural: read on the positive-denominator representative gives in , and a nonnegative integer is the image of a unique natural.
The numerators sum to in : applying scaling to the real sequence multiplies the hypothesis by to give , and in the field by multiplicativity of the embedding applied to the rational identity , so .
The partial sums of the numerators, inside : define by recursion, and for (and for ), so each is natural and . Its canonical field image is for , the sum being real under the convention in [L5]. Indeed the identity holds at zero; additivity of the embedding gives , and the real finite-sum recursion then gives the successor identity.
Hence as natural numbers: steps 2.1 and 2.2 give , and the map is injective on . This licenses the natural list length N in the next step; it is not an identification of arbitrary reals with indices.
The expanded list: define by whenever for some , and for ; this covers every exactly once, because the blocks for partition by step 3.1, and every with is positive since each .
Its sum and product, by a second induction written out rather than asserted. The claim at is and . At we have and all four expressions are the empty sum or the empty product . Assume the claim at . Since , splitting gives , and the tail is by definition , since ; every one of its terms equals , because for , so the constant sum evaluates it as , and therefore by the recursion clause for finite sums. The same computation with products in place of sums, the constant product in place of the constant sum, gives . At , where by step 3.1, this reads and .
Applying AM-GM to and substituting: , the last equality by scaling together with .
Rewriting the left-hand side with the rational power laws: , each being positive.
Combining the two displays gives , which is the assertion.
Depends on
The arithmetic mean, geometric mean inequality
Rational powers of a positive base
Laws of rational exponents
Finite sums and finite products, by recursion
Laws of finite sums and finite products
The principle of mathematical induction
The recursion theorem
Every rational has a positive-denominator representative
The rationals as equivalence classes of pairs of integers
Arithmetic on the rationals
Order on the rationals
Order on the integers
The naturals embed in the integers
The unique embedding of ℚ into an ordered field
Integer powers
Existence and uniqueness of -th roots: a unique with
Canonical naturals are positive and strictly increasing