Source release: Oxford examination archive. The cover dates this examination to 17 June 2011, 9.30–11.00am. The unlettered requests in Question 1 are covered together in its parent Solution, with separate Solutions for its two numbered interval tests. Questions 2 and 3 each have one complete Solution.
A unital Banach star-algebra is a complex unital Banach algebra with a continuous conjugate-linear algebra involution satisfying
The algebra norm is complete and submultiplicative, with the unit conventionally normalized to norm one. Many definitions additionally require the involution to be isometric, ; the function algebra here satisfies this stronger convention as well. A Banach star-algebra does not by definition require the extra C-star identity .
Here is the lattice proof of the Stone-Weierstrass theorem in its complex self-adjoint form. Let be the uniform closure of the given unital star-subalgebra in , and let be its real-valued part. Closure preserves multiplication and complex conjugation, so is still a unital star-algebra. Real and imaginary parts lie in , and separation by the complex algebra implies separation by its real-valued part.
For , polynomial approximation of absolute value on a bounded interval gives . For example, this polynomial approximation follows from the binomial series for the square root, whose coefficients give uniform convergence on the relevant closed interval. Thus
also lie in . It is therefore a lattice of functions.
To see how the lattice gives density, fix a real continuous target and a tolerance . For any two distinct points , separation and constants provide with and , by an affine rescaling of a separating function. For , use the constant .
Fix . The sets where cover as varies. A finite subcover and a pointwise maximum give with everywhere and . The neighborhoods where now cover as varies. A second finite subcover and a pointwise minimum give satisfying
This is lattice approximation from two-point approximation. Using half the desired tolerance gives strict uniform error below the desired bound. Since is closed, it contains every real continuous function. Splitting a complex function into real and imaginary parts proves .
The annulus test. For the closed annulus, the family holomorphic in the interior is not dense in the full continuous-function algebra. A uniform limit on the annulus is locally uniform in its interior, so locally uniform convergence of holomorphic functions makes that limit holomorphic there. Thus the proposed family is closed in . It is proper because is continuous but not holomorphic on the annulus interior.
There is also an explicit continuous functional detecting the obstruction. Choose a small closed disk of radius contained in the annulus interior and let be the integral of around its positively oriented boundary. The Cauchy integral theorem gives for every member of the holomorphic family, while parametrizing the circle gives . Since , no uniform approximation of by that family is possible.
The even-power family is dense on . It is the algebra . It contains constants, is closed under complex conjugation, and separates points because the real function is injective on this positive interval. The complex Stone-Weierstrass theorem therefore applies.
Equivalently, the squaring map is a homeomorphism from onto . Given , approximate the continuous function uniformly on by complex polynomials, using the Weierstrass approximation theorem on the real and imaginary parts. Substitution produces exactly the required even-power approximants. The positive interval is what lets even powers distinguish all its points.
The odd-power family is also dense on . It is , where is the even-power family in part (i). Multiplication by is an invertible bounded linear map on , with inverse division by , since .
For an explicit approximation, take any and . The quotient is continuous, so part (i) gives an even polynomial with . Then belongs to the odd-power family and
Although this family is not a subalgebra and does not contain the constant function exactly, it can approximate every continuous function, including constants. This proves even and odd polynomial density on a positive interval without incorrectly applying the unital algebra theorem directly to the odd family.
An element is a Normal element of a C-star algebra if . It is a Unitary element of a C-star algebra if . Its Banach algebra spectrum is
where is the group of invertible elements of a Banach algebra.
For normal , the continuous functional calculus is the unique unital isometric star-isomorphism
which sends the coordinate function to . It preserves sums, products, complex scalars and the involution: . In particular . The spectral mapping theorem in this setting states
The common unit and spectral permanence for C-star subalgebras permit computing this spectrum in itself.
For unitary , the C-star identity gives . If , has a multiplicative inverse by the Neumann series. If , write and apply the same series. Therefore lies on the unit circle.
For , the denominator is nonzero because . The function is consequently continuous on the spectrum, and
Thus throughout , so there. Applying the star-homomorphism proves
Hence is unitary. Explicitly, the same calculus gives
with the denominator's inverse also supplied by a norm-convergent Neumann series. This is the Möbius transform of a unitary element. The conjugation on in the denominator is essential to the unit-modulus identity.
A state on a C-star algebra is a positive linear functional of norm one. In the unital case this is equivalent to positivity and . A unital star-homomorphism is a cyclic representation of a C-star algebra if it has a vector with . For the representation associated with a state, the cyclic vector is taken to have norm one.
Use inner products linear in the first argument, as in the displayed formula for the projection in the source. In the Gelfand–Naimark–Segal construction, set
The Cauchy–Schwarz inequality for positive C-star functionals makes this inner product well-defined on . Indeed a null vector pairs to zero with every vector. The set is a left ideal: implies
so implies . The same estimate shows that left multiplication induces a bounded operator
Complete the quotient pre-Hilbert space to , and extend these operators by continuity. The formula for the inner product gives , while multiplication and the unit are preserved. Put . Then , the vectors are dense in , and
This gives the requested cyclic representation for an arbitrary state.
For the specified unit vector , the functional is complex-linear and obeys . Moreover
Thus it is positive and has norm exactly one, proving that is a vector state on a C-star algebra.
The rank-one projection satisfies
If , this is zero for every , so . Conversely, evaluation at gives . Hence
In particular .
The concrete GNS space is the completion of with
Define . The description of the null ideal proves it is well-defined, and the displayed equality proves it is a linear isometry with range . It therefore extends to an isometry . The extended range is closed, since its domain is complete, and contains the dense set . Thus is onto the required closure.
In fact here , so : given any , the bounded rank-one operator sends to . Thus is a unitary map onto all of , and .
Finally, on the dense quotient vectors,
Both sides are bounded operators, so this identity extends to all of . Multiplication by the inverse unitary map gives
This is the GNS representation of a vector state, with the vector state, null ideal, cyclic vector and intertwining isometry all made explicit.

Articles by others on the same topic (0)

There are currently no matching articles.