Let GG be a Lie group and let πV:GGL(V)\pi_V:G\to\operatorname{GL}(V) and πW:GGL(W)\pi_W:G\to\operatorname{GL}(W) be . Their tensor product representation on VWV\otimes W is

πVW(g)=πV(g)πW(g).\pi_{V\otimes W}(g)=\pi_V(g)\otimes\pi_W(g).

Thus

πVW(g)(vw)=πV(g)vπW(g)w,\pi_{V\otimes W}(g)(v\otimes w) =\pi_V(g)v\otimes\pi_W(g)w,

extended linearly from pure tensors.

Lie-algebra version

Let g\mathfrak g be a Lie algebra, and let ρV:ggl(V)\rho_V:\mathfrak g\to\mathfrak{gl}(V) and ρW:ggl(W)\rho_W:\mathfrak g\to\mathfrak{gl}(W) be . The tensor product representation ρVW:ggl(VW)\rho_{V\otimes W}:\mathfrak g\to\mathfrak{gl}(V\otimes W) is

ρVW(X)=ρV(X)IdW+IdVρW(X),\rho_{V\otimes W}(X) =\rho_V(X)\otimes\operatorname{Id}_W +\operatorname{Id}_V\otimes\rho_W(X),

or, equivalently,

X(vw)=(Xv)w+v(Xw).X\cdot (v\otimes w)=(X\cdot v)\otimes w + v\otimes (X\cdot w).

A direct computation using the commutator bracket on gl(VW)\mathfrak{gl}(V\otimes W) shows that ρVW\rho_{V\otimes W} is a .

Weight behavior

If g\mathfrak g is semisimple and h\mathfrak h is a , then tensor products interact cleanly with the : if vVλv\in V_\lambda and wWμw\in W_\mu, then

vw(VW)λ+μ.v\otimes w \in (V\otimes W)_{\lambda+\mu}.

Thus the set of of VWV\otimes W is contained in the Minkowski sum of the weight sets of VV and WW. This underlies Clebsch–Gordan decompositions and highest-weight calculations.