It's clear that the chromatic number of $Q_n$ is $2$. But what about the graph $G$ with vertex set ${n}^{(r)}$ where two vertices are adjacent if and only if their coordiantes differ by one?
Can't seem to find any sort of good colouring at all here.
It's clear that the chromatic number of $Q_n$ is $2$. But what about the graph $G$ with vertex set ${n}^{(r)}$ where two vertices are adjacent if and only if their coordiantes differ by one?
Can't seem to find any sort of good colouring at all here.
Copyright © 2021 JogjaFile Inc.
The chromatic number is $n.$
Indeed, consider that the vertices are $r$-tuples from $\{1,\ldots,n\}.$ Coloring the vertex $v = (v_1,\ldots,v_r)$ with color $$c(v) = \sum_{i=1}^r v_i \pmod{n},$$ gives you a proper coloring. For if $$v = (v_1,\ldots,v_{i-1},x,v_{i+1},\ldots,v_{r}) \quad \mbox{and} \quad v' = (v_1,\ldots,v_{i-1},y,v_{i+1},\ldots,v_{r})$$ are two distinct adjacent vertices then $v$ receives the color $$c(v) = c(v') - x + y \pmod{n} \ne c(v')$$ since $x,y \leq n.$
To see that the chromatic number is at least $n$ note that given fixed elements $v_1,\ldots,v_{n-1}$ the set $\{ (v_1,\ldots,v_{n-1},k) \mid k \in \{1,\ldots,n\}\}$ induces a $n$-clique in $G.$
Another more direct way to see this is that your graphs are the cartesian products of $K_n$ that is $G_r = \square_{i=1}^r K_n$ and a well known identity says $$\chi(G \square H) = \rm{max}(\chi(G), \chi(H)).$$