![]() |
Metamath
Proof Explorer Theorem List (p. 422 of 426) | < Previous Next > |
Bad symbols? Try the
GIF version. |
||
Mirrors > Metamath Home Page > MPE Home Page > Theorem List Contents > Recent Proofs This page: Page List |
Color key: | ![]() (1-27775) |
![]() (27776-29300) |
![]() (29301-42551) |
Type | Label | Description |
---|---|---|
Statement | ||
Theorem | fldcALTV 42101* | The restriction of the category of division rings to the set of field homomorphisms is a category, the "category of fields". (Contributed by AV, 20-Feb-2020.) (New usage is discouraged.) |
⊢ 𝐶 = (𝑈 ∩ DivRing) & ⊢ 𝐽 = (𝑟 ∈ 𝐶, 𝑠 ∈ 𝐶 ↦ (𝑟 RingHom 𝑠)) & ⊢ 𝐷 = (𝑈 ∩ Field) & ⊢ 𝐹 = (𝑟 ∈ 𝐷, 𝑠 ∈ 𝐷 ↦ (𝑟 RingHom 𝑠)) ⇒ ⊢ (𝑈 ∈ 𝑉 → (((RingCatALTV‘𝑈) ↾cat 𝐽) ↾cat 𝐹) ∈ Cat) | ||
Theorem | fldhmsubcALTV 42102* | According to df-subc 16472, the subcategories (Subcat‘𝐶) of a category 𝐶 are subsets of the homomorphisms of 𝐶 ( see subcssc 16500 and subcss2 16503). Therefore, the set of field homomorphisms is a "subcategory" of the category of division rings. (Contributed by AV, 20-Feb-2020.) (New usage is discouraged.) |
⊢ 𝐶 = (𝑈 ∩ DivRing) & ⊢ 𝐽 = (𝑟 ∈ 𝐶, 𝑠 ∈ 𝐶 ↦ (𝑟 RingHom 𝑠)) & ⊢ 𝐷 = (𝑈 ∩ Field) & ⊢ 𝐹 = (𝑟 ∈ 𝐷, 𝑠 ∈ 𝐷 ↦ (𝑟 RingHom 𝑠)) ⇒ ⊢ (𝑈 ∈ 𝑉 → 𝐹 ∈ (Subcat‘((RingCatALTV‘𝑈) ↾cat 𝐽))) | ||
Theorem | rngcrescrhmALTV 42103 | The category of non-unital rings (in a universe) restricted to the ring homomorphisms between unital rings (in the same universe). (Contributed by AV, 1-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ (𝜑 → (𝐶 ↾cat 𝐻) = ((𝐶 ↾s 𝑅) sSet 〈(Hom ‘ndx), 𝐻〉)) | ||
Theorem | rhmsubcALTVlem1 42104 | Lemma 1 for rhmsubcALTV 42108. (Contributed by AV, 2-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ (𝜑 → 𝐻 Fn (𝑅 × 𝑅)) | ||
Theorem | rhmsubcALTVlem2 42105 | Lemma 2 for rhmsubcALTV 42108. (Contributed by AV, 2-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ ((𝜑 ∧ 𝑋 ∈ 𝑅 ∧ 𝑌 ∈ 𝑅) → (𝑋𝐻𝑌) = (𝑋 RingHom 𝑌)) | ||
Theorem | rhmsubcALTVlem3 42106* | Lemma 3 for rhmsubcALTV 42108. (Contributed by AV, 2-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ ((𝜑 ∧ 𝑥 ∈ 𝑅) → ((Id‘(RngCatALTV‘𝑈))‘𝑥) ∈ (𝑥𝐻𝑥)) | ||
Theorem | rhmsubcALTVlem4 42107* | Lemma 4 for rhmsubcALTV 42108. (Contributed by AV, 2-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ ((((𝜑 ∧ 𝑥 ∈ 𝑅) ∧ (𝑦 ∈ 𝑅 ∧ 𝑧 ∈ 𝑅)) ∧ (𝑓 ∈ (𝑥𝐻𝑦) ∧ 𝑔 ∈ (𝑦𝐻𝑧))) → (𝑔(〈𝑥, 𝑦〉(comp‘(RngCatALTV‘𝑈))𝑧)𝑓) ∈ (𝑥𝐻𝑧)) | ||
Theorem | rhmsubcALTV 42108 | According to df-subc 16472, the subcategories (Subcat‘𝐶) of a category 𝐶 are subsets of the homomorphisms of 𝐶 ( see subcssc 16500 and subcss2 16503). Therefore, the set of unital ring homomorphisms is a "subcategory" of the category of non-unital rings. (Contributed by AV, 2-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ (𝜑 → 𝐻 ∈ (Subcat‘(RngCatALTV‘𝑈))) | ||
Theorem | rhmsubcALTVcat 42109 | The restriction of the category of non-unital rings to the set of unital ring homomorphisms is a category. (Contributed by AV, 4-Mar-2020.) (New usage is discouraged.) |
⊢ (𝜑 → 𝑈 ∈ 𝑉) & ⊢ 𝐶 = (RngCatALTV‘𝑈) & ⊢ (𝜑 → 𝑅 = (Ring ∩ 𝑈)) & ⊢ 𝐻 = ( RingHom ↾ (𝑅 × 𝑅)) ⇒ ⊢ (𝜑 → ((RngCatALTV‘𝑈) ↾cat 𝐻) ∈ Cat) | ||
Theorem | xpprsng 42110 | The Cartesian product of an unordered pair and a singleton. (Contributed by AV, 20-May-2019.) |
⊢ ((𝐴 ∈ 𝑉 ∧ 𝐵 ∈ 𝑊 ∧ 𝐶 ∈ 𝑈) → ({𝐴, 𝐵} × {𝐶}) = {〈𝐴, 𝐶〉, 〈𝐵, 𝐶〉}) | ||
Theorem | opeliun2xp 42111 | Membership of an ordered pair in a union of Cartesian products over its second component, analogous to opeliunxp 5170. (Contributed by AV, 30-Mar-2019.) |
⊢ (〈𝐶, 𝑦〉 ∈ ∪ 𝑦 ∈ 𝐵 (𝐴 × {𝑦}) ↔ (𝑦 ∈ 𝐵 ∧ 𝐶 ∈ 𝐴)) | ||
Theorem | eliunxp2 42112* | Membership in a union of Cartesian products over its second component, analogous to eliunxp 5259. (Contributed by AV, 30-Mar-2019.) |
⊢ (𝐶 ∈ ∪ 𝑦 ∈ 𝐵 (𝐴 × {𝑦}) ↔ ∃𝑥∃𝑦(𝐶 = 〈𝑥, 𝑦〉 ∧ (𝑥 ∈ 𝐴 ∧ 𝑦 ∈ 𝐵))) | ||
Theorem | mpt2mptx2 42113* | Express a two-argument function as a one-argument function, or vice-versa. In this version 𝐴(𝑦) is not assumed to be constant w.r.t 𝑦, analogous to mpt2mptx 6751. (Contributed by AV, 30-Mar-2019.) |
⊢ (𝑧 = 〈𝑥, 𝑦〉 → 𝐶 = 𝐷) ⇒ ⊢ (𝑧 ∈ ∪ 𝑦 ∈ 𝐵 (𝐴 × {𝑦}) ↦ 𝐶) = (𝑥 ∈ 𝐴, 𝑦 ∈ 𝐵 ↦ 𝐷) | ||
Theorem | cbvmpt2x2 42114* | Rule to change the bound variable in a maps-to function, using implicit substitution. This version of cbvmpt2 6734 allows 𝐴 to be a function of 𝑦, analogous to cbvmpt2x 6733. (Contributed by AV, 30-Mar-2019.) |
⊢ Ⅎ𝑧𝐴 & ⊢ Ⅎ𝑦𝐷 & ⊢ Ⅎ𝑧𝐶 & ⊢ Ⅎ𝑤𝐶 & ⊢ Ⅎ𝑥𝐸 & ⊢ Ⅎ𝑦𝐸 & ⊢ (𝑦 = 𝑧 → 𝐴 = 𝐷) & ⊢ ((𝑦 = 𝑧 ∧ 𝑥 = 𝑤) → 𝐶 = 𝐸) ⇒ ⊢ (𝑥 ∈ 𝐴, 𝑦 ∈ 𝐵 ↦ 𝐶) = (𝑤 ∈ 𝐷, 𝑧 ∈ 𝐵 ↦ 𝐸) | ||
Theorem | dmmpt2ssx2 42115* | The domain of a mapping is a subset of its base classes expressed as union of Cartesian products over its second component, analogous to dmmpt2ssx 7235. (Contributed by AV, 30-Mar-2019.) |
⊢ 𝐹 = (𝑥 ∈ 𝐴, 𝑦 ∈ 𝐵 ↦ 𝐶) ⇒ ⊢ dom 𝐹 ⊆ ∪ 𝑦 ∈ 𝐵 (𝐴 × {𝑦}) | ||
Theorem | mpt2exxg2 42116* | Existence of an operation class abstraction (version for dependent domains, i.e. the first base class may depend on the second base class), analogous to mpt2exxg 7244. (Contributed by AV, 30-Mar-2019.) |
⊢ 𝐹 = (𝑥 ∈ 𝐴, 𝑦 ∈ 𝐵 ↦ 𝐶) ⇒ ⊢ ((𝐵 ∈ 𝑅 ∧ ∀𝑦 ∈ 𝐵 𝐴 ∈ 𝑆) → 𝐹 ∈ V) | ||
Theorem | ovmpt2rdxf 42117* | Value of an operation given by a maps-to rule, deduction form, with substitution of second argument, analogous to ovmpt2dxf 6786. (Contributed by AV, 30-Mar-2019.) |
⊢ (𝜑 → 𝐹 = (𝑥 ∈ 𝐶, 𝑦 ∈ 𝐷 ↦ 𝑅)) & ⊢ ((𝜑 ∧ (𝑥 = 𝐴 ∧ 𝑦 = 𝐵)) → 𝑅 = 𝑆) & ⊢ ((𝜑 ∧ 𝑦 = 𝐵) → 𝐶 = 𝐿) & ⊢ (𝜑 → 𝐴 ∈ 𝐿) & ⊢ (𝜑 → 𝐵 ∈ 𝐷) & ⊢ (𝜑 → 𝑆 ∈ 𝑋) & ⊢ Ⅎ𝑥𝜑 & ⊢ Ⅎ𝑦𝜑 & ⊢ Ⅎ𝑦𝐴 & ⊢ Ⅎ𝑥𝐵 & ⊢ Ⅎ𝑥𝑆 & ⊢ Ⅎ𝑦𝑆 ⇒ ⊢ (𝜑 → (𝐴𝐹𝐵) = 𝑆) | ||
Theorem | ovmpt2rdx 42118* | Value of an operation given by a maps-to rule, deduction form, with substitution of second argument, analogous to ovmpt2dxf 6786. (Contributed by AV, 30-Mar-2019.) |
⊢ (𝜑 → 𝐹 = (𝑥 ∈ 𝐶, 𝑦 ∈ 𝐷 ↦ 𝑅)) & ⊢ ((𝜑 ∧ (𝑥 = 𝐴 ∧ 𝑦 = 𝐵)) → 𝑅 = 𝑆) & ⊢ ((𝜑 ∧ 𝑦 = 𝐵) → 𝐶 = 𝐿) & ⊢ (𝜑 → 𝐴 ∈ 𝐿) & ⊢ (𝜑 → 𝐵 ∈ 𝐷) & ⊢ (𝜑 → 𝑆 ∈ 𝑋) ⇒ ⊢ (𝜑 → (𝐴𝐹𝐵) = 𝑆) | ||
Theorem | ovmpt2x2 42119* | The value of an operation class abstraction. Variant of ovmpt2ga 6790 which does not require 𝐷 and 𝑥 to be distinct. (Contributed by Jeff Madsen, 10-Jun-2010.) (Revised by Mario Carneiro, 20-Dec-2013.) |
⊢ ((𝑥 = 𝐴 ∧ 𝑦 = 𝐵) → 𝑅 = 𝑆) & ⊢ (𝑦 = 𝐵 → 𝐶 = 𝐿) & ⊢ 𝐹 = (𝑥 ∈ 𝐶, 𝑦 ∈ 𝐷 ↦ 𝑅) ⇒ ⊢ ((𝐴 ∈ 𝐿 ∧ 𝐵 ∈ 𝐷 ∧ 𝑆 ∈ 𝐻) → (𝐴𝐹𝐵) = 𝑆) | ||
Theorem | fdmdifeqresdif 42120* | The restriction of a conditional mapping to function values of a function having a domain which is a difference with a singleton equals this function. (Contributed by AV, 23-Apr-2019.) |
⊢ 𝐹 = (𝑥 ∈ 𝐷 ↦ if(𝑥 = 𝑌, 𝑋, (𝐺‘𝑥))) ⇒ ⊢ (𝐺:(𝐷 ∖ {𝑌})⟶𝑅 → 𝐺 = (𝐹 ↾ (𝐷 ∖ {𝑌}))) | ||
Theorem | offvalfv 42121* | The function operation expressed as a mapping with function values. (Contributed by AV, 6-Apr-2019.) |
⊢ (𝜑 → 𝐴 ∈ 𝑉) & ⊢ (𝜑 → 𝐹 Fn 𝐴) & ⊢ (𝜑 → 𝐺 Fn 𝐴) ⇒ ⊢ (𝜑 → (𝐹 ∘𝑓 𝑅𝐺) = (𝑥 ∈ 𝐴 ↦ ((𝐹‘𝑥)𝑅(𝐺‘𝑥)))) | ||
Theorem | ofaddmndmap 42122 | The function operation applied to the addition for functions (with the same domain) into a monoid is a function (with the same domain) into the monoid. (Contributed by AV, 6-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) & ⊢ + = (+g‘𝑀) ⇒ ⊢ ((𝑀 ∈ Mnd ∧ 𝑉 ∈ 𝑌 ∧ (𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ 𝐵 ∈ (𝑅 ↑𝑚 𝑉))) → (𝐴 ∘𝑓 + 𝐵) ∈ (𝑅 ↑𝑚 𝑉)) | ||
Theorem | mapsnop 42123 | A singleton of an ordered pair as an element of the mapping operation. (Contributed by AV, 12-Apr-2019.) |
⊢ 𝐹 = {〈𝑋, 𝑌〉} ⇒ ⊢ ((𝑋 ∈ 𝑉 ∧ 𝑌 ∈ 𝑅 ∧ 𝑅 ∈ 𝑊) → 𝐹 ∈ (𝑅 ↑𝑚 {𝑋})) | ||
Theorem | mapprop 42124 | An unordered pair containing two ordered pairs as an element of the mapping operation. (Contributed by AV, 16-Apr-2019.) |
⊢ 𝐹 = {〈𝑋, 𝐴〉, 〈𝑌, 𝐵〉} ⇒ ⊢ (((𝑋 ∈ 𝑉 ∧ 𝐴 ∈ 𝑅) ∧ (𝑌 ∈ 𝑉 ∧ 𝐵 ∈ 𝑅) ∧ (𝑋 ≠ 𝑌 ∧ 𝑅 ∈ 𝑊)) → 𝐹 ∈ (𝑅 ↑𝑚 {𝑋, 𝑌})) | ||
Theorem | ztprmneprm 42125 | A prime is not an integer multiple of another prime. (Contributed by AV, 23-May-2019.) |
⊢ ((𝑍 ∈ ℤ ∧ 𝐴 ∈ ℙ ∧ 𝐵 ∈ ℙ) → ((𝑍 · 𝐴) = 𝐵 → 𝐴 = 𝐵)) | ||
Theorem | 2t6m3t4e0 42126 | 2 times 6 minus 3 times 4 equals 0. (Contributed by AV, 24-May-2019.) |
⊢ ((2 · 6) − (3 · 4)) = 0 | ||
Theorem | ssnn0ssfz 42127* | For any finite subset of ℕ0, find a superset in the form of a set of sequential integers, analogous to ssnnssfz 29549. (Contributed by AV, 30-Sep-2019.) |
⊢ (𝐴 ∈ (𝒫 ℕ0 ∩ Fin) → ∃𝑛 ∈ ℕ0 𝐴 ⊆ (0...𝑛)) | ||
Theorem | nn0sumltlt 42128 | If the sum of two nonnegative integers is less than a third integer, then one of the summands is already less than this third integer. (Contributed by AV, 19-Oct-2019.) |
⊢ ((𝑎 ∈ ℕ0 ∧ 𝑏 ∈ ℕ0 ∧ 𝑐 ∈ ℕ0) → ((𝑎 + 𝑏) < 𝑐 → 𝑏 < 𝑐)) | ||
Theorem | bcpascm1 42129 | Pascal's rule for the binomial coefficient, generalized to all integers 𝐾, shifted down by 1. (Contributed by AV, 8-Sep-2019.) |
⊢ ((𝑁 ∈ ℕ ∧ 𝐾 ∈ ℤ) → (((𝑁 − 1)C𝐾) + ((𝑁 − 1)C(𝐾 − 1))) = (𝑁C𝐾)) | ||
Theorem | altgsumbc 42130* | The sum of binomial coefficients for a fixed positive 𝑁 with alternating signs is zero. Notice that this is not valid for 𝑁 = 0 (since ((-1↑0) · (0C0)) = (1 · 1) = 1). For a proof using Pascal's rule (bcpascm1 42129) instead of the binomial theorem (binom 14562) , see altgsumbcALT 42131. (Contributed by AV, 13-Sep-2019.) |
⊢ (𝑁 ∈ ℕ → Σ𝑘 ∈ (0...𝑁)((-1↑𝑘) · (𝑁C𝑘)) = 0) | ||
Theorem | altgsumbcALT 42131* | Alternate proof of altgsumbc 42130, using Pascal's rule (bcpascm1 42129) instead of the binomial theorem (binom 14562). (Contributed by AV, 8-Sep-2019.) (Proof modification is discouraged.) (New usage is discouraged.) |
⊢ (𝑁 ∈ ℕ → Σ𝑘 ∈ (0...𝑁)((-1↑𝑘) · (𝑁C𝑘)) = 0) | ||
Theorem | zlmodzxzlmod 42132 | The ℤ-module ℤ × ℤ is a (left) module with the ring of integers as base set. (Contributed by AV, 20-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) ⇒ ⊢ (𝑍 ∈ LMod ∧ ℤring = (Scalar‘𝑍)) | ||
Theorem | zlmodzxzel 42133 | An element of the (base set of the) ℤ-module ℤ × ℤ. (Contributed by AV, 21-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) ⇒ ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ) → {〈0, 𝐴〉, 〈1, 𝐵〉} ∈ (Base‘𝑍)) | ||
Theorem | zlmodzxz0 42134 | The 0 of the ℤ-module ℤ × ℤ. (Contributed by AV, 20-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) & ⊢ 0 = {〈0, 0〉, 〈1, 0〉} ⇒ ⊢ 0 = (0g‘𝑍) | ||
Theorem | zlmodzxzscm 42135 | The scalar multiplication of the ℤ-module ℤ × ℤ. (Contributed by AV, 20-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) & ⊢ ∙ = ( ·𝑠 ‘𝑍) ⇒ ⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ ∧ 𝐶 ∈ ℤ) → (𝐴 ∙ {〈0, 𝐵〉, 〈1, 𝐶〉}) = {〈0, (𝐴 · 𝐵)〉, 〈1, (𝐴 · 𝐶)〉}) | ||
Theorem | zlmodzxzadd 42136 | The addition of the ℤ-module ℤ × ℤ. (Contributed by AV, 22-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) & ⊢ + = (+g‘𝑍) ⇒ ⊢ (((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ) ∧ (𝐶 ∈ ℤ ∧ 𝐷 ∈ ℤ)) → ({〈0, 𝐴〉, 〈1, 𝐶〉} + {〈0, 𝐵〉, 〈1, 𝐷〉}) = {〈0, (𝐴 + 𝐵)〉, 〈1, (𝐶 + 𝐷)〉}) | ||
Theorem | zlmodzxzsubm 42137 | The subtraction of the ℤ-module ℤ × ℤ expressed as addition. (Contributed by AV, 24-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) & ⊢ − = (-g‘𝑍) ⇒ ⊢ (((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ) ∧ (𝐶 ∈ ℤ ∧ 𝐷 ∈ ℤ)) → ({〈0, 𝐴〉, 〈1, 𝐶〉} − {〈0, 𝐵〉, 〈1, 𝐷〉}) = ({〈0, 𝐴〉, 〈1, 𝐶〉} (+g‘𝑍)(-1( ·𝑠 ‘𝑍){〈0, 𝐵〉, 〈1, 𝐷〉}))) | ||
Theorem | zlmodzxzsub 42138 | The subtraction of the ℤ-module ℤ × ℤ. (Contributed by AV, 22-May-2019.) (Revised by AV, 10-Jun-2019.) |
⊢ 𝑍 = (ℤring freeLMod {0, 1}) & ⊢ − = (-g‘𝑍) ⇒ ⊢ (((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ) ∧ (𝐶 ∈ ℤ ∧ 𝐷 ∈ ℤ)) → ({〈0, 𝐴〉, 〈1, 𝐶〉} − {〈0, 𝐵〉, 〈1, 𝐷〉}) = {〈0, (𝐴 − 𝐵)〉, 〈1, (𝐶 − 𝐷)〉}) | ||
Theorem | gsumpr 42139* | Group sum of a pair. (Contributed by AV, 6-Dec-2018.) (Proof shortened by AV, 28-Jul-2019.) |
⊢ 𝐵 = (Base‘𝐺) & ⊢ + = (+g‘𝐺) & ⊢ (𝑘 = 𝑀 → 𝐴 = 𝐶) & ⊢ (𝑘 = 𝑁 → 𝐴 = 𝐷) ⇒ ⊢ ((𝐺 ∈ CMnd ∧ (𝑀 ∈ 𝑉 ∧ 𝑁 ∈ 𝑊 ∧ 𝑀 ≠ 𝑁) ∧ (𝐶 ∈ 𝐵 ∧ 𝐷 ∈ 𝐵)) → (𝐺 Σg (𝑘 ∈ {𝑀, 𝑁} ↦ 𝐴)) = (𝐶 + 𝐷)) | ||
Theorem | mgpsumunsn 42140* | Extract a summand/factor from the group sum for the multiplicative group of a unital ring. (Contributed by AV, 29-Dec-2018.) |
⊢ 𝑀 = (mulGrp‘𝑅) & ⊢ · = (.r‘𝑅) & ⊢ (𝜑 → 𝑅 ∈ CRing) & ⊢ (𝜑 → 𝑁 ∈ Fin) & ⊢ (𝜑 → 𝐼 ∈ 𝑁) & ⊢ ((𝜑 ∧ 𝑘 ∈ 𝑁) → 𝐴 ∈ (Base‘𝑅)) & ⊢ (𝜑 → 𝑋 ∈ (Base‘𝑅)) & ⊢ (𝑘 = 𝐼 → 𝐴 = 𝑋) ⇒ ⊢ (𝜑 → (𝑀 Σg (𝑘 ∈ 𝑁 ↦ 𝐴)) = ((𝑀 Σg (𝑘 ∈ (𝑁 ∖ {𝐼}) ↦ 𝐴)) · 𝑋)) | ||
Theorem | mgpsumz 42141* | If the group sum for the multiplicative group of a unital ring contains a summand/factor that is the zero of the ring, the group sum itself is zero. (Contributed by AV, 29-Dec-2018.) |
⊢ 𝑀 = (mulGrp‘𝑅) & ⊢ · = (.r‘𝑅) & ⊢ (𝜑 → 𝑅 ∈ CRing) & ⊢ (𝜑 → 𝑁 ∈ Fin) & ⊢ (𝜑 → 𝐼 ∈ 𝑁) & ⊢ ((𝜑 ∧ 𝑘 ∈ 𝑁) → 𝐴 ∈ (Base‘𝑅)) & ⊢ 0 = (0g‘𝑅) & ⊢ (𝑘 = 𝐼 → 𝐴 = 0 ) ⇒ ⊢ (𝜑 → (𝑀 Σg (𝑘 ∈ 𝑁 ↦ 𝐴)) = 0 ) | ||
Theorem | mgpsumn 42142* | If the group sum for the multiplicative group of a unital ring contains a summand/factor that is the one of the ring, this summand/ factor can be removed from the group sum. (Contributed by AV, 29-Dec-2018.) |
⊢ 𝑀 = (mulGrp‘𝑅) & ⊢ · = (.r‘𝑅) & ⊢ (𝜑 → 𝑅 ∈ CRing) & ⊢ (𝜑 → 𝑁 ∈ Fin) & ⊢ (𝜑 → 𝐼 ∈ 𝑁) & ⊢ ((𝜑 ∧ 𝑘 ∈ 𝑁) → 𝐴 ∈ (Base‘𝑅)) & ⊢ 1 = (1r‘𝑅) & ⊢ (𝑘 = 𝐼 → 𝐴 = 1 ) ⇒ ⊢ (𝜑 → (𝑀 Σg (𝑘 ∈ 𝑁 ↦ 𝐴)) = (𝑀 Σg (𝑘 ∈ (𝑁 ∖ {𝐼}) ↦ 𝐴))) | ||
Theorem | gsumsplit2f 42143* | Split a group sum into two parts. (Contributed by AV, 4-Sep-2019.) |
⊢ Ⅎ𝑘𝜑 & ⊢ 𝐵 = (Base‘𝐺) & ⊢ 0 = (0g‘𝐺) & ⊢ + = (+g‘𝐺) & ⊢ (𝜑 → 𝐺 ∈ CMnd) & ⊢ (𝜑 → 𝐴 ∈ 𝑉) & ⊢ ((𝜑 ∧ 𝑘 ∈ 𝐴) → 𝑋 ∈ 𝐵) & ⊢ (𝜑 → (𝑘 ∈ 𝐴 ↦ 𝑋) finSupp 0 ) & ⊢ (𝜑 → (𝐶 ∩ 𝐷) = ∅) & ⊢ (𝜑 → 𝐴 = (𝐶 ∪ 𝐷)) ⇒ ⊢ (𝜑 → (𝐺 Σg (𝑘 ∈ 𝐴 ↦ 𝑋)) = ((𝐺 Σg (𝑘 ∈ 𝐶 ↦ 𝑋)) + (𝐺 Σg (𝑘 ∈ 𝐷 ↦ 𝑋)))) | ||
Theorem | gsumdifsndf 42144* | Extract a summand from a finitely supported group sum. (Contributed by AV, 4-Sep-2019.) |
⊢ Ⅎ𝑘𝑌 & ⊢ Ⅎ𝑘𝜑 & ⊢ 𝐵 = (Base‘𝐺) & ⊢ + = (+g‘𝐺) & ⊢ (𝜑 → 𝐺 ∈ CMnd) & ⊢ (𝜑 → 𝐴 ∈ 𝑊) & ⊢ (𝜑 → (𝑘 ∈ 𝐴 ↦ 𝑋) finSupp (0g‘𝐺)) & ⊢ ((𝜑 ∧ 𝑘 ∈ 𝐴) → 𝑋 ∈ 𝐵) & ⊢ (𝜑 → 𝑀 ∈ 𝐴) & ⊢ (𝜑 → 𝑌 ∈ 𝐵) & ⊢ ((𝜑 ∧ 𝑘 = 𝑀) → 𝑋 = 𝑌) ⇒ ⊢ (𝜑 → (𝐺 Σg (𝑘 ∈ 𝐴 ↦ 𝑋)) = ((𝐺 Σg (𝑘 ∈ (𝐴 ∖ {𝑀}) ↦ 𝑋)) + 𝑌)) | ||
Theorem | exple2lt6 42145 | A nonnegative integer to the power of itself is less than 6 if it is less than or equal to 2. (Contributed by AV, 16-Mar-2019.) |
⊢ ((𝑁 ∈ ℕ0 ∧ 𝑁 ≤ 2) → (𝑁↑𝑁) < 6) | ||
Theorem | pgrple2abl 42146 | Every symmetric group on a set with at most 2 elements is abelian. (Contributed by AV, 16-Mar-2019.) |
⊢ 𝐺 = (SymGrp‘𝐴) ⇒ ⊢ ((𝐴 ∈ 𝑉 ∧ (#‘𝐴) ≤ 2) → 𝐺 ∈ Abel) | ||
Theorem | pgrpgt2nabl 42147 | Every symmetric group on a set with more than 2 elements is not abelian, see also the remark in [Rotman] p. 28. (Contributed by AV, 21-Mar-2019.) |
⊢ 𝐺 = (SymGrp‘𝐴) ⇒ ⊢ ((𝐴 ∈ 𝑉 ∧ 2 < (#‘𝐴)) → 𝐺 ∉ Abel) | ||
Theorem | invginvrid 42148 | Identity for a multiplication with additive and multiplicative inverses in a ring. (Contributed by AV, 18-May-2018.) |
⊢ 𝐵 = (Base‘𝑅) & ⊢ 𝑈 = (Unit‘𝑅) & ⊢ 𝑁 = (invg‘𝑅) & ⊢ 𝐼 = (invr‘𝑅) & ⊢ · = (.r‘𝑅) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝑋 ∈ 𝐵 ∧ 𝑌 ∈ 𝑈) → ((𝑁‘𝑌) · ((𝐼‘(𝑁‘𝑌)) · 𝑋)) = 𝑋) | ||
Theorem | rmsupp0 42149* | The support of a mapping of a multiplication of zero with a function into a ring is empty. (Contributed by AV, 10-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Ring ∧ 𝑉 ∈ 𝑋 ∧ 𝐶 = (0g‘𝑀)) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉)) → ((𝑣 ∈ 𝑉 ↦ (𝐶(.r‘𝑀)(𝐴‘𝑣))) supp (0g‘𝑀)) = ∅) | ||
Theorem | domnmsuppn0 42150* | The support of a mapping of a multiplication of a nonzero constant with a function into a (ring theoretic) domain equals the support of the function. (Contributed by AV, 11-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Domn ∧ 𝑉 ∈ 𝑋) ∧ (𝐶 ∈ 𝑅 ∧ 𝐶 ≠ (0g‘𝑀)) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉)) → ((𝑣 ∈ 𝑉 ↦ (𝐶(.r‘𝑀)(𝐴‘𝑣))) supp (0g‘𝑀)) = (𝐴 supp (0g‘𝑀))) | ||
Theorem | rmsuppss 42151* | The support of a mapping of a multiplication of a constant with a function into a ring is a subset of the support of the function. (Contributed by AV, 11-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Ring ∧ 𝑉 ∈ 𝑋 ∧ 𝐶 ∈ 𝑅) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉)) → ((𝑣 ∈ 𝑉 ↦ (𝐶(.r‘𝑀)(𝐴‘𝑣))) supp (0g‘𝑀)) ⊆ (𝐴 supp (0g‘𝑀))) | ||
Theorem | mndpsuppss 42152 | The support of a mapping of a scalar multiplication with a function of scalars is a subset of the support of the function of scalars. (Contributed by AV, 5-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Mnd ∧ 𝑉 ∈ 𝑋) ∧ (𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ 𝐵 ∈ (𝑅 ↑𝑚 𝑉))) → ((𝐴 ∘𝑓 (+g‘𝑀)𝐵) supp (0g‘𝑀)) ⊆ ((𝐴 supp (0g‘𝑀)) ∪ (𝐵 supp (0g‘𝑀)))) | ||
Theorem | scmsuppss 42153* | The support of a mapping of a scalar multiplication with a function of scalars is a subset of the support of the function of scalars. (Contributed by AV, 5-Apr-2019.) |
⊢ 𝑆 = (Scalar‘𝑀) & ⊢ 𝑅 = (Base‘𝑆) ⇒ ⊢ ((𝑀 ∈ LMod ∧ 𝑉 ∈ 𝒫 (Base‘𝑀) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉)) → ((𝑣 ∈ 𝑉 ↦ ((𝐴‘𝑣)( ·𝑠 ‘𝑀)𝑣)) supp (0g‘𝑀)) ⊆ (𝐴 supp (0g‘𝑆))) | ||
Theorem | rmsuppfi 42154* | The support of a mapping of a multiplication of a constant with a function into a ring is finite if the support of the function is finite. (Contributed by AV, 11-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Ring ∧ 𝑉 ∈ 𝑋 ∧ 𝐶 ∈ 𝑅) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ (𝐴 supp (0g‘𝑀)) ∈ Fin) → ((𝑣 ∈ 𝑉 ↦ (𝐶(.r‘𝑀)(𝐴‘𝑣))) supp (0g‘𝑀)) ∈ Fin) | ||
Theorem | rmfsupp 42155* | A mapping of a multiplication of a constant with a function into a ring is finitely supported if the function is finitely supported. (Contributed by AV, 9-Jun-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Ring ∧ 𝑉 ∈ 𝑋 ∧ 𝐶 ∈ 𝑅) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ 𝐴 finSupp (0g‘𝑀)) → (𝑣 ∈ 𝑉 ↦ (𝐶(.r‘𝑀)(𝐴‘𝑣))) finSupp (0g‘𝑀)) | ||
Theorem | mndpsuppfi 42156 | The support of a mapping of a scalar multiplication with a function of scalars is finite if the support of the function of scalars is finite. (Contributed by AV, 5-Apr-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Mnd ∧ 𝑉 ∈ 𝑋) ∧ (𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ 𝐵 ∈ (𝑅 ↑𝑚 𝑉)) ∧ ((𝐴 supp (0g‘𝑀)) ∈ Fin ∧ (𝐵 supp (0g‘𝑀)) ∈ Fin)) → ((𝐴 ∘𝑓 (+g‘𝑀)𝐵) supp (0g‘𝑀)) ∈ Fin) | ||
Theorem | mndpfsupp 42157 | A mapping of a scalar multiplication with a function of scalars is finitely supported if the function of scalars is finitely supported. (Contributed by AV, 9-Jun-2019.) |
⊢ 𝑅 = (Base‘𝑀) ⇒ ⊢ (((𝑀 ∈ Mnd ∧ 𝑉 ∈ 𝑋) ∧ (𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ 𝐵 ∈ (𝑅 ↑𝑚 𝑉)) ∧ (𝐴 finSupp (0g‘𝑀) ∧ 𝐵 finSupp (0g‘𝑀))) → (𝐴 ∘𝑓 (+g‘𝑀)𝐵) finSupp (0g‘𝑀)) | ||
Theorem | scmsuppfi 42158* | The support of a mapping of a scalar multiplication with a function of scalars is finite if the support of the function of scalars is finite. (Contributed by AV, 5-Apr-2019.) |
⊢ 𝑆 = (Scalar‘𝑀) & ⊢ 𝑅 = (Base‘𝑆) ⇒ ⊢ (((𝑀 ∈ LMod ∧ 𝑉 ∈ 𝒫 (Base‘𝑀)) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ (𝐴 supp (0g‘𝑆)) ∈ Fin) → ((𝑣 ∈ 𝑉 ↦ ((𝐴‘𝑣)( ·𝑠 ‘𝑀)𝑣)) supp (0g‘𝑀)) ∈ Fin) | ||
Theorem | scmfsupp 42159* | A mapping of a scalar multiplication with a function of scalars is finitely supported if the function of scalars is finitely supported. (Contributed by AV, 9-Jun-2019.) |
⊢ 𝑆 = (Scalar‘𝑀) & ⊢ 𝑅 = (Base‘𝑆) ⇒ ⊢ (((𝑀 ∈ LMod ∧ 𝑉 ∈ 𝒫 (Base‘𝑀)) ∧ 𝐴 ∈ (𝑅 ↑𝑚 𝑉) ∧ 𝐴 finSupp (0g‘𝑆)) → (𝑣 ∈ 𝑉 ↦ ((𝐴‘𝑣)( ·𝑠 ‘𝑀)𝑣)) finSupp (0g‘𝑀)) | ||
Theorem | suppmptcfin 42160* | The support of a mapping with value 0 except of one is finite. (Contributed by AV, 27-Apr-2019.) |
⊢ 𝐵 = (Base‘𝑀) & ⊢ 𝑅 = (Scalar‘𝑀) & ⊢ 0 = (0g‘𝑅) & ⊢ 1 = (1r‘𝑅) & ⊢ 𝐹 = (𝑥 ∈ 𝑉 ↦ if(𝑥 = 𝑋, 1 , 0 )) ⇒ ⊢ ((𝑀 ∈ LMod ∧ 𝑉 ∈ 𝒫 𝐵 ∧ 𝑋 ∈ 𝑉) → (𝐹 supp 0 ) ∈ Fin) | ||
Theorem | mptcfsupp 42161* | A mapping with value 0 except of one is finitely supported. (Contributed by AV, 9-Jun-2019.) |
⊢ 𝐵 = (Base‘𝑀) & ⊢ 𝑅 = (Scalar‘𝑀) & ⊢ 0 = (0g‘𝑅) & ⊢ 1 = (1r‘𝑅) & ⊢ 𝐹 = (𝑥 ∈ 𝑉 ↦ if(𝑥 = 𝑋, 1 , 0 )) ⇒ ⊢ ((𝑀 ∈ LMod ∧ 𝑉 ∈ 𝒫 𝐵 ∧ 𝑋 ∈ 𝑉) → 𝐹 finSupp 0 ) | ||
Theorem | fsuppmptdmf 42162* | A mapping with a finite domain is finitely supported. (Contributed by AV, 4-Sep-2019.) |
⊢ Ⅎ𝑥𝜑 & ⊢ 𝐹 = (𝑥 ∈ 𝐴 ↦ 𝑌) & ⊢ (𝜑 → 𝐴 ∈ Fin) & ⊢ ((𝜑 ∧ 𝑥 ∈ 𝐴) → 𝑌 ∈ 𝑉) & ⊢ (𝜑 → 𝑍 ∈ 𝑊) ⇒ ⊢ (𝜑 → 𝐹 finSupp 𝑍) | ||
Theorem | lmodvsmdi 42163 | Multiple distributive law for scalar product (left-distributivity). (Contributed by AV, 5-Sep-2019.) |
⊢ 𝑉 = (Base‘𝑊) & ⊢ 𝐹 = (Scalar‘𝑊) & ⊢ · = ( ·𝑠 ‘𝑊) & ⊢ 𝐾 = (Base‘𝐹) & ⊢ ↑ = (.g‘𝑊) & ⊢ 𝐸 = (.g‘𝐹) ⇒ ⊢ ((𝑊 ∈ LMod ∧ (𝑅 ∈ 𝐾 ∧ 𝑁 ∈ ℕ0 ∧ 𝑋 ∈ 𝑉)) → (𝑅 · (𝑁 ↑ 𝑋)) = ((𝑁𝐸𝑅) · 𝑋)) | ||
Theorem | gsumlsscl 42164* | Closure of a group sum in a linear subspace: A (finitely supported) sum of scalar multiplications of vectors of a subset of a linear subspace is also contained in the linear subspace. (Contributed by AV, 20-Apr-2019.) (Revised by AV, 28-Jul-2019.) |
⊢ 𝑆 = (LSubSp‘𝑀) & ⊢ 𝑅 = (Scalar‘𝑀) & ⊢ 𝐵 = (Base‘𝑅) ⇒ ⊢ ((𝑀 ∈ LMod ∧ 𝑍 ∈ 𝑆 ∧ 𝑉 ⊆ 𝑍) → ((𝐹 ∈ (𝐵 ↑𝑚 𝑉) ∧ 𝐹 finSupp (0g‘𝑅)) → (𝑀 Σg (𝑣 ∈ 𝑉 ↦ ((𝐹‘𝑣)( ·𝑠 ‘𝑀)𝑣))) ∈ 𝑍)) | ||
Theorem | ascl0 42165 | The scalar 0 embedded into a left module corresponds to the 0 of the left module if the left module is also a ring. (Contributed by AV, 31-Jul-2019.) |
⊢ 𝐴 = (algSc‘𝑊) & ⊢ 𝐹 = (Scalar‘𝑊) & ⊢ (𝜑 → 𝑊 ∈ LMod) & ⊢ (𝜑 → 𝑊 ∈ Ring) ⇒ ⊢ (𝜑 → (𝐴‘(0g‘𝐹)) = (0g‘𝑊)) | ||
Theorem | ascl1 42166 | The scalar 1 embedded into a left module corresponds to the 1 of the left module if the left module is also a ring. (Contributed by AV, 31-Jul-2019.) |
⊢ 𝐴 = (algSc‘𝑊) & ⊢ 𝐹 = (Scalar‘𝑊) & ⊢ (𝜑 → 𝑊 ∈ LMod) & ⊢ (𝜑 → 𝑊 ∈ Ring) ⇒ ⊢ (𝜑 → (𝐴‘(1r‘𝐹)) = (1r‘𝑊)) | ||
Theorem | assaascl0 42167 | The scalar 0 embedded into an associative algebra corresponds to the 0 of the associative algebra. (Contributed by AV, 31-Jul-2019.) |
⊢ 𝐴 = (algSc‘𝑊) & ⊢ 𝐹 = (Scalar‘𝑊) & ⊢ (𝜑 → 𝑊 ∈ AssAlg) ⇒ ⊢ (𝜑 → (𝐴‘(0g‘𝐹)) = (0g‘𝑊)) | ||
Theorem | assaascl1 42168 | The scalar 1 embedded into an associative algebra corresponds to the 1 of the an associative algebra. (Contributed by AV, 31-Jul-2019.) |
⊢ 𝐴 = (algSc‘𝑊) & ⊢ 𝐹 = (Scalar‘𝑊) & ⊢ (𝜑 → 𝑊 ∈ AssAlg) ⇒ ⊢ (𝜑 → (𝐴‘(1r‘𝐹)) = (1r‘𝑊)) | ||
Theorem | ply1vr1smo 42169 | The variable in a polynomial expressed as scaled monomial. (Contributed by AV, 12-Aug-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 1 = (1r‘𝑅) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ 𝐺 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝐺) & ⊢ 𝑋 = (var1‘𝑅) ⇒ ⊢ (𝑅 ∈ Ring → ( 1 · (1 ↑ 𝑋)) = 𝑋) | ||
Theorem | ply1ass23l 42170 | Associative identity with scalar and ring multiplication for the polynomial ring. (Contributed by AV, 14-Aug-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ × = (.r‘𝑃) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐾 = (Base‘𝑅) & ⊢ · = ( ·𝑠 ‘𝑃) ⇒ ⊢ ((𝑅 ∈ Ring ∧ (𝐴 ∈ 𝐾 ∧ 𝑋 ∈ 𝐵 ∧ 𝑌 ∈ 𝐵)) → ((𝐴 · 𝑋) × 𝑌) = (𝐴 · (𝑋 × 𝑌))) | ||
Theorem | ply1sclrmsm 42171 | The ring multiplication of a polynomial with a scalar polynomial is equal to the scalar multiplication of the polynomial with the corresponding scalar. (Contributed by AV, 14-Aug-2019.) |
⊢ 𝐾 = (Base‘𝑅) & ⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐸 = (Base‘𝑃) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ × = (.r‘𝑃) & ⊢ 𝑁 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝑁) & ⊢ 𝐴 = (algSc‘𝑃) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝐹 ∈ 𝐾 ∧ 𝑍 ∈ 𝐸) → ((𝐴‘𝐹) × 𝑍) = (𝐹 · 𝑍)) | ||
Theorem | coe1id 42172* | Coefficient vector of the unit polynomial. (Contributed by AV, 9-Aug-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐼 = (1r‘𝑃) & ⊢ 0 = (0g‘𝑅) & ⊢ 1 = (1r‘𝑅) ⇒ ⊢ (𝑅 ∈ Ring → (coe1‘𝐼) = (𝑥 ∈ ℕ0 ↦ if(𝑥 = 0, 1 , 0 ))) | ||
Theorem | coe1sclmulval 42173 | The value of the coefficient vector of a polynomial multiplied on the left by a scalar. (Contributed by AV, 14-Aug-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐾 = (Base‘𝑅) & ⊢ 𝐴 = (algSc‘𝑃) & ⊢ 𝑆 = ( ·𝑠 ‘𝑃) & ⊢ ∙ = (.r‘𝑃) & ⊢ · = (.r‘𝑅) ⇒ ⊢ ((𝑅 ∈ Ring ∧ (𝑌 ∈ 𝐾 ∧ 𝑍 ∈ 𝐵) ∧ 𝑁 ∈ ℕ0) → ((coe1‘(𝑌𝑆𝑍))‘𝑁) = (𝑌 · ((coe1‘𝑍)‘𝑁))) | ||
Theorem | ply1mulgsumlem1 42174* | Lemma 1 for ply1mulgsum 42178. (Contributed by AV, 19-Oct-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐴 = (coe1‘𝐾) & ⊢ 𝐶 = (coe1‘𝐿) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ × = (.r‘𝑃) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ ∗ = (.r‘𝑅) & ⊢ 𝑀 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝑀) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝐾 ∈ 𝐵 ∧ 𝐿 ∈ 𝐵) → ∃𝑠 ∈ ℕ0 ∀𝑛 ∈ ℕ0 (𝑠 < 𝑛 → ((𝐴‘𝑛) = (0g‘𝑅) ∧ (𝐶‘𝑛) = (0g‘𝑅)))) | ||
Theorem | ply1mulgsumlem2 42175* | Lemma 2 for ply1mulgsum 42178. (Contributed by AV, 19-Oct-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐴 = (coe1‘𝐾) & ⊢ 𝐶 = (coe1‘𝐿) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ × = (.r‘𝑃) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ ∗ = (.r‘𝑅) & ⊢ 𝑀 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝑀) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝐾 ∈ 𝐵 ∧ 𝐿 ∈ 𝐵) → ∃𝑠 ∈ ℕ0 ∀𝑛 ∈ ℕ0 (𝑠 < 𝑛 → (𝑅 Σg (𝑙 ∈ (0...𝑛) ↦ ((𝐴‘𝑙) ∗ (𝐶‘(𝑛 − 𝑙))))) = (0g‘𝑅))) | ||
Theorem | ply1mulgsumlem3 42176* | Lemma 3 for ply1mulgsum 42178. (Contributed by AV, 20-Oct-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐴 = (coe1‘𝐾) & ⊢ 𝐶 = (coe1‘𝐿) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ × = (.r‘𝑃) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ ∗ = (.r‘𝑅) & ⊢ 𝑀 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝑀) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝐾 ∈ 𝐵 ∧ 𝐿 ∈ 𝐵) → (𝑘 ∈ ℕ0 ↦ (𝑅 Σg (𝑙 ∈ (0...𝑘) ↦ ((𝐴‘𝑙) ∗ (𝐶‘(𝑘 − 𝑙)))))) finSupp (0g‘𝑅)) | ||
Theorem | ply1mulgsumlem4 42177* | Lemma 4 for ply1mulgsum 42178. (Contributed by AV, 19-Oct-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐴 = (coe1‘𝐾) & ⊢ 𝐶 = (coe1‘𝐿) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ × = (.r‘𝑃) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ ∗ = (.r‘𝑅) & ⊢ 𝑀 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝑀) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝐾 ∈ 𝐵 ∧ 𝐿 ∈ 𝐵) → (𝑘 ∈ ℕ0 ↦ ((𝑅 Σg (𝑙 ∈ (0...𝑘) ↦ ((𝐴‘𝑙) ∗ (𝐶‘(𝑘 − 𝑙))))) · (𝑘 ↑ 𝑋))) finSupp (0g‘𝑃)) | ||
Theorem | ply1mulgsum 42178* | The product of two polynomials expressed as group sum of scaled monomials. (Contributed by AV, 20-Oct-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐴 = (coe1‘𝐾) & ⊢ 𝐶 = (coe1‘𝐿) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ × = (.r‘𝑃) & ⊢ · = ( ·𝑠 ‘𝑃) & ⊢ ∗ = (.r‘𝑅) & ⊢ 𝑀 = (mulGrp‘𝑃) & ⊢ ↑ = (.g‘𝑀) ⇒ ⊢ ((𝑅 ∈ Ring ∧ 𝐾 ∈ 𝐵 ∧ 𝐿 ∈ 𝐵) → (𝐾 × 𝐿) = (𝑃 Σg (𝑘 ∈ ℕ0 ↦ ((𝑅 Σg (𝑙 ∈ (0...𝑘) ↦ ((𝐴‘𝑙) ∗ (𝐶‘(𝑘 − 𝑙))))) · (𝑘 ↑ 𝑋))))) | ||
Theorem | evl1at0 42179 | Polynomial evaluation for the 0 scalar. (Contributed by AV, 10-Aug-2019.) |
⊢ 𝑂 = (eval1‘𝑅) & ⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 0 = (0g‘𝑅) & ⊢ 𝑍 = (0g‘𝑃) ⇒ ⊢ (𝑅 ∈ CRing → ((𝑂‘𝑍)‘ 0 ) = 0 ) | ||
Theorem | evl1at1 42180 | Polynomial evaluation for the 1 scalar. (Contributed by AV, 10-Aug-2019.) |
⊢ 𝑂 = (eval1‘𝑅) & ⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 1 = (1r‘𝑅) & ⊢ 𝐼 = (1r‘𝑃) ⇒ ⊢ (𝑅 ∈ CRing → ((𝑂‘𝐼)‘ 1 ) = 1 ) | ||
Theorem | linply1 42181 | A term of the form 𝑥 − 𝐶 is a (univariate) polynomial, also called "linear polynomial". (Part of ply1remlem 23922). (Contributed by AV, 3-Jul-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐾 = (Base‘𝑅) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ − = (-g‘𝑃) & ⊢ 𝐴 = (algSc‘𝑃) & ⊢ 𝐺 = (𝑋 − (𝐴‘𝐶)) & ⊢ (𝜑 → 𝐶 ∈ 𝐾) & ⊢ (𝜑 → 𝑅 ∈ Ring) ⇒ ⊢ (𝜑 → 𝐺 ∈ 𝐵) | ||
Theorem | lineval 42182 | A term of the form 𝑥 − 𝐶 evaluated for 𝑥 = 𝑉 results in 𝑉 − 𝐶 (part of ply1remlem 23922). (Contributed by AV, 3-Jul-2019.) |
⊢ 𝑃 = (Poly1‘𝑅) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝐾 = (Base‘𝑅) & ⊢ 𝑋 = (var1‘𝑅) & ⊢ − = (-g‘𝑃) & ⊢ 𝐴 = (algSc‘𝑃) & ⊢ 𝐺 = (𝑋 − (𝐴‘𝐶)) & ⊢ (𝜑 → 𝐶 ∈ 𝐾) & ⊢ 𝑂 = (eval1‘𝑅) & ⊢ (𝜑 → 𝑅 ∈ CRing) & ⊢ (𝜑 → 𝑉 ∈ 𝐾) ⇒ ⊢ (𝜑 → ((𝑂‘𝐺)‘𝑉) = (𝑉(-g‘𝑅)𝐶)) | ||
Theorem | zringsubgval 42183 | Subtraction in the ring of integers. (Contributed by AV, 3-Aug-2019.) |
⊢ − = (-g‘ℤring) ⇒ ⊢ ((𝑋 ∈ ℤ ∧ 𝑌 ∈ ℤ) → (𝑋 − 𝑌) = (𝑋 − 𝑌)) | ||
Theorem | linevalexample 42184 | The polynomial 𝑥 − 3 over ℤ evaluated for 𝑥 = 5 results in 2. (Contributed by AV, 3-Jul-2019.) |
⊢ 𝑃 = (Poly1‘ℤring) & ⊢ 𝐵 = (Base‘𝑃) & ⊢ 𝑋 = (var1‘ℤring) & ⊢ − = (-g‘𝑃) & ⊢ 𝐴 = (algSc‘𝑃) & ⊢ 𝐺 = (𝑋 − (𝐴‘3)) & ⊢ 𝑂 = (eval1‘ℤring) ⇒ ⊢ ((𝑂‘(𝑋 − (𝐴‘3)))‘5) = 2 | ||
In the following, alternative definitions for diagonal and scalar matrices are provided. These definitions define diagonal and scalar matrices as extensible structures, whereas the definitions df-dmat 20296 and df-scmat 20297 define diagonal and scalar matrices as sets. | ||
Syntax | cdmatalt 42185 | Alternative notation for the algebra of diagonal matrices. |
class DMatALT | ||
Syntax | cscmatalt 42186 | Alternative notation for the algebra of scalar matrices. |
class ScMatALT | ||
Definition | df-dmatalt 42187* | Define the set of n x n diagonal (square) matrices over a set (usually a ring) r, see definition in [Roman] p. 4 or Definition 3.12 in [Hefferon] p. 240. (Contributed by AV, 8-Dec-2019.) |
⊢ DMatALT = (𝑛 ∈ Fin, 𝑟 ∈ V ↦ ⦋(𝑛 Mat 𝑟) / 𝑎⦌(𝑎 ↾s {𝑚 ∈ (Base‘𝑎) ∣ ∀𝑖 ∈ 𝑛 ∀𝑗 ∈ 𝑛 (𝑖 ≠ 𝑗 → (𝑖𝑚𝑗) = (0g‘𝑟))})) | ||
Definition | df-scmatalt 42188* | Define the algebra of n x n scalar matrices over a set (usually a ring) r, see definition in [Connell] p. 57: "A scalar matrix is a diagonal matrix for which all the diagonal terms are equal, i.e., a matrix of the form cIn". (Contributed by AV, 8-Dec-2019.) |
⊢ ScMatALT = (𝑛 ∈ Fin, 𝑟 ∈ V ↦ ⦋(𝑛 Mat 𝑟) / 𝑎⦌(𝑎 ↾s {𝑚 ∈ (Base‘𝑎) ∣ ∃𝑐 ∈ (Base‘𝑟)∀𝑖 ∈ 𝑛 ∀𝑗 ∈ 𝑛 (𝑖𝑚𝑗) = if(𝑖 = 𝑗, 𝑐, (0g‘𝑟))})) | ||
Theorem | dmatALTval 42189* | The algebra of 𝑁 x 𝑁 diagonal matrices over a ring 𝑅. (Contributed by AV, 8-Dec-2019.) |
⊢ 𝐴 = (𝑁 Mat 𝑅) & ⊢ 𝐵 = (Base‘𝐴) & ⊢ 0 = (0g‘𝑅) & ⊢ 𝐷 = (𝑁 DMatALT 𝑅) ⇒ ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ V) → 𝐷 = (𝐴 ↾s {𝑚 ∈ 𝐵 ∣ ∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖 ≠ 𝑗 → (𝑖𝑚𝑗) = 0 )})) | ||
Theorem | dmatALTbas 42190* | The base set of the algebra of 𝑁 x 𝑁 diagonal matrices over a ring 𝑅, i.e. the set of all 𝑁 x 𝑁 diagonal matrices over the ring 𝑅. (Contributed by AV, 8-Dec-2019.) |
⊢ 𝐴 = (𝑁 Mat 𝑅) & ⊢ 𝐵 = (Base‘𝐴) & ⊢ 0 = (0g‘𝑅) & ⊢ 𝐷 = (𝑁 DMatALT 𝑅) ⇒ ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ V) → (Base‘𝐷) = {𝑚 ∈ 𝐵 ∣ ∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖 ≠ 𝑗 → (𝑖𝑚𝑗) = 0 )}) | ||
Theorem | dmatALTbasel 42191* | An element of the base set of the algebra of 𝑁 x 𝑁 diagonal matrices over a ring 𝑅, i.e. an 𝑁 x 𝑁 diagonal matrix over the ring 𝑅. (Contributed by AV, 8-Dec-2019.) |
⊢ 𝐴 = (𝑁 Mat 𝑅) & ⊢ 𝐵 = (Base‘𝐴) & ⊢ 0 = (0g‘𝑅) & ⊢ 𝐷 = (𝑁 DMatALT 𝑅) ⇒ ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ V) → (𝑀 ∈ (Base‘𝐷) ↔ (𝑀 ∈ 𝐵 ∧ ∀𝑖 ∈ 𝑁 ∀𝑗 ∈ 𝑁 (𝑖 ≠ 𝑗 → (𝑖𝑀𝑗) = 0 )))) | ||
Theorem | dmatbas 42192 | The set of all 𝑁 x 𝑁 diagonal matrices over (the ring) 𝑅 is the base set of the algebra of 𝑁 x 𝑁 diagonal matrices over (the ring) 𝑅. (Contributed by AV, 8-Dec-2019.) |
⊢ 𝐴 = (𝑁 Mat 𝑅) & ⊢ 𝐵 = (Base‘𝐴) & ⊢ 0 = (0g‘𝑅) & ⊢ 𝐷 = (𝑁 DMat 𝑅) ⇒ ⊢ ((𝑁 ∈ Fin ∧ 𝑅 ∈ 𝑉) → 𝐷 = (Base‘(𝑁 DMatALT 𝑅))) | ||
According to Wikipedia ("Linear combination", 29-Mar-2019,
https://en.wikipedia.org/wiki/Linear_combination) "In mathematics, a
linear combination is an expression constructed from a set of terms by
multiplying each term by a constant and adding the results (e.g., a linear
combination of x and y would be any expression of the form ax + by, where a
and b are constants). The concept of linear combinations is central to
linear algebra and related fields of mathematics." In linear algebra, these
"terms" are "vectors" (elements from vector spaces or left modules), and the
constants are elements of the underlying field resp. ring. This corresponds
to the definition in [Lang] p. 129: "Let M be a module over a ring A and let
S be a subset of M. By a linear combination of elements of S (with
coefficients in A) one means a sum ∑x ∈S
axx where {ax} is a set of elements of A, ...". In the
definition in [Lang] p. 129, it is additionally claimed that "..., almost all
of which [elements of A] are equal to 0.". This is not necessarily required
in the following definition df-linc 42195, but it is essential if additions and
scalar multiplications of linear combinations are considered. Therefore, we
define the set of all linear combinations with finite support in df-lco 42196,
so that we can show that such sets are submodules of the corresponding
modules, see lincolss 42223.
| ||
Syntax | clinc 42193 | Extend class notation with the operation constructing a linear combination (of vectors from a left module). |
class linC | ||
Syntax | clinco 42194 | Extend class notation with the operation constructing a set of linear combinations (of vectors from a left module) with finite support. |
class LinCo | ||
Definition | df-linc 42195* | Define the operation constructing a linear combination. Although this definition is taylored for linear combinations of vectors from left modules, it can be used for any structure having a Base, Scalar s and a scalar multiplication ·𝑠. (Contributed by AV, 29-Mar-2019.) |
⊢ linC = (𝑚 ∈ V ↦ (𝑠 ∈ ((Base‘(Scalar‘𝑚)) ↑𝑚 𝑣), 𝑣 ∈ 𝒫 (Base‘𝑚) ↦ (𝑚 Σg (𝑥 ∈ 𝑣 ↦ ((𝑠‘𝑥)( ·𝑠 ‘𝑚)𝑥))))) | ||
Definition | df-lco 42196* | Define the operation constructing the set of all linear combinations for a set of vectors. (Contributed by AV, 31-Mar-2019.) (Revised by AV, 28-Jul-2019.) |
⊢ LinCo = (𝑚 ∈ V, 𝑣 ∈ 𝒫 (Base‘𝑚) ↦ {𝑐 ∈ (Base‘𝑚) ∣ ∃𝑠 ∈ ((Base‘(Scalar‘𝑚)) ↑𝑚 𝑣)(𝑠 finSupp (0g‘(Scalar‘𝑚)) ∧ 𝑐 = (𝑠( linC ‘𝑚)𝑣))}) | ||
Theorem | lincop 42197* | A linear combination as operation. (Contributed by AV, 30-Mar-2019.) |
⊢ (𝑀 ∈ 𝑋 → ( linC ‘𝑀) = (𝑠 ∈ ((Base‘(Scalar‘𝑀)) ↑𝑚 𝑣), 𝑣 ∈ 𝒫 (Base‘𝑀) ↦ (𝑀 Σg (𝑥 ∈ 𝑣 ↦ ((𝑠‘𝑥)( ·𝑠 ‘𝑀)𝑥))))) | ||
Theorem | lincval 42198* | The value of a linear combination. (Contributed by AV, 30-Mar-2019.) |
⊢ ((𝑀 ∈ 𝑋 ∧ 𝑆 ∈ ((Base‘(Scalar‘𝑀)) ↑𝑚 𝑉) ∧ 𝑉 ∈ 𝒫 (Base‘𝑀)) → (𝑆( linC ‘𝑀)𝑉) = (𝑀 Σg (𝑥 ∈ 𝑉 ↦ ((𝑆‘𝑥)( ·𝑠 ‘𝑀)𝑥)))) | ||
Theorem | dflinc2 42199* | Alternative definition of linear combinations using the function operation. (Contributed by AV, 1-Apr-2019.) |
⊢ linC = (𝑚 ∈ V ↦ (𝑠 ∈ ((Base‘(Scalar‘𝑚)) ↑𝑚 𝑣), 𝑣 ∈ 𝒫 (Base‘𝑚) ↦ (𝑚 Σg (𝑠 ∘𝑓 ( ·𝑠 ‘𝑚)( I ↾ 𝑣))))) | ||
Theorem | lcoop 42200* | A linear combination as operation. (Contributed by AV, 5-Apr-2019.) (Revised by AV, 28-Jul-2019.) |
⊢ 𝐵 = (Base‘𝑀) & ⊢ 𝑆 = (Scalar‘𝑀) & ⊢ 𝑅 = (Base‘𝑆) ⇒ ⊢ ((𝑀 ∈ 𝑋 ∧ 𝑉 ∈ 𝒫 𝐵) → (𝑀 LinCo 𝑉) = {𝑐 ∈ 𝐵 ∣ ∃𝑠 ∈ (𝑅 ↑𝑚 𝑉)(𝑠 finSupp (0g‘𝑆) ∧ 𝑐 = (𝑠( linC ‘𝑀)𝑉))}) |
< Previous Next > |
Copyright terms: Public domain | < Previous Next > |