TheoremBase

Convex Combinations: Small Cases, Splitting Off the Last Point, Dropping a Zero Weight, Joining Two Combinations, and Membership in Convex Sets

Convex combinations of one and two points, the reduction of a combination of N+1 points to a segment point between a combination of N points and the last point, removal of a point of weight zero, the combination obtained by joining two combinations, and closure of convex sets under convex combinations.

Statement

In the setting of Convex Sets in Rn\mathbb{R}^n: Standing Notation, with the standard simplices ΔN\Delta_{N} and convex combinations ∑k=1Ntkxk\sum_{k=1}^{N}t_{k}x_{k}.

One point. A point t∈R1t\in\mathbb{R}^{1} lies in Δ1\Delta_{1} if and only if t1=1t_{1}=1. For x1∈Rnx_{1}\in\mathbb{R}^{n} and t∈Δ1t\in\Delta_{1},

∑k=11tkxk=x1.\sum_{k=1}^{1}t_{k}x_{k}=x_{1}.

Two points. For x1,x2∈Rnx_{1},x_{2}\in\mathbb{R}^{n}, a point t∈R2t\in\mathbb{R}^{2} lies in Δ2\Delta_{2} if and only if t=(1−s,s)t=(1-s,s) for some s∈[0,1]s\in[0,1], and then

∑k=12tkxk=(1−s)x1+sx2.\sum_{k=1}^{2}t_{k}x_{k}=(1-s)x_{1}+sx_{2}.

Splitting off the last point. Let N∈NN\in\mathbb{N}, x1,…,xN+1∈Rnx_{1},\dots,x_{N+1}\in\mathbb{R}^{n}, t∈ΔN+1t\in\Delta_{N+1} and z=∑k=1N+1tkxkz=\sum_{k=1}^{N+1}t_{k}x_{k}. Then 0≤tN+1≤10\le t_{N+1}\le1. If tN+1=1t_{N+1}=1, then z=xN+1z=x_{N+1}. If tN+1<1t_{N+1}<1, then s=1−tN+1>0s=1-t_{N+1}>0, the point τ=(t1/s,…,tN/s)\tau=(t_{1}/s,\dots,t_{N}/s) lies in ΔN\Delta_{N}, and, with y=∑k=1Nτkxky=\sum_{k=1}^{N}\tau_{k}x_{k},

z=(1−tN+1) y+tN+1 xN+1.z=(1-t_{N+1})\,y+t_{N+1}\,x_{N+1}.

Dropping a zero weight. Let N∈NN\in\mathbb{N}, x1,…,xN+1∈Rnx_{1},\dots,x_{N+1}\in\mathbb{R}^{n}, t∈ΔN+1t\in\Delta_{N+1} and j∈[N+1]j\in[N+1] with tj=0t_{j}=0, and let gj:[N]→[N+1]∖{j}g_{j}:[N]\to[N+1]\setminus\{j\} be the bijection of Bijections between Intervals: the Gap Map and the Reversal (i). Then (tgj(1),…,tgj(N))∈ΔN(t_{g_{j}(1)},\dots,t_{g_{j}(N)})\in\Delta_{N} and

∑k=1N+1tkxk=∑k=1Ntgj(k)xgj(k).\sum_{k=1}^{N+1}t_{k}x_{k}=\sum_{k=1}^{N}t_{g_{j}(k)}x_{g_{j}(k)}.

Joining two combinations. Let N,P∈NN,P\in\mathbb{N}, x1,…,xN∈Rnx_{1},\dots,x_{N}\in\mathbb{R}^{n}, x1′,…,xP′∈Rnx'_{1},\dots,x'_{P}\in\mathbb{R}^{n}, τ∈ΔN\tau\in\Delta_{N}, σ∈ΔP\sigma\in\Delta_{P} and λ∈[0,1]\lambda\in[0,1]. Let w1,…,wN+Pw_{1},\dots,w_{N+P} be the points with wk=xkw_{k}=x_{k} for k∈[N]k\in[N] and wN+j=xj′w_{N+j}=x'_{j} for j∈[P]j\in[P], and let t∈RN+Pt\in\mathbb{R}^{N+P} have tk=(1−λ)τkt_{k}=(1-\lambda)\tau_{k} for k∈[N]k\in[N] and tN+j=λσjt_{N+j}=\lambda\sigma_{j} for j∈[P]j\in[P]. Then t∈ΔN+Pt\in\Delta_{N+P} and

(1−λ)∑k=1Nτkxk+λ∑j=1Pσjxj′=∑k=1N+Ptkwk.(1-\lambda)\sum_{k=1}^{N}\tau_{k}x_{k}+\lambda\sum_{j=1}^{P}\sigma_{j}x'_{j}=\sum_{k=1}^{N+P}t_{k}w_{k}.

Membership. Let C⊆RnC\subseteq\mathbb{R}^{n} be convex, N∈NN\in\mathbb{N}, x1,…,xN∈Cx_{1},\dots,x_{N}\in C and t∈ΔNt\in\Delta_{N}. Then ∑k=1Ntkxk∈C\sum_{k=1}^{N}t_{k}x_{k}\in C.

Proofs

Log in to submit a proof.

Loading...

Citations

Loading…

Dependencies

Loading…

Related

0 relations

Curated associations between results. These are editable and subjective — they do not replace the dependency graph, which is derived from the references in the text.

No relations recorded yet.

Comments

Log in to comment.

Loading…