Ensembles, Purification & the HJW Theorem
Different mixtures, same density matrix, and why it matters.
The same density matrix can be built from infinitely many ensembles, and no experiment can tell them apart. The Hughston–Jozsa–Wootters theorem says exactly how these ensembles are related, and shows that whoever holds a purification of your system can remotely choose which ensemble you have, without ever changing your .
What you’ll learn
- State and prove the unitary freedom in the ensemble for a density operator.
- Construct purifications and prove that they are unique up to an isometry on the purifying system.
- Prove the Schrödinger–HJW steering theorem and derive no-signalling from it.
- Characterize which pure states, with which weights, can appear in an ensemble for a given (support condition, , majorization).
Unitary freedom in the ensemble
Work with unnormalized vectors , so that an ensemble is just a list of vectors with , and the weights are recovered as . Pad shorter lists with zero vectors whenever two lists are compared.
(⇐) Direct computation, using :
(⇒) Let with , and set . Every lies in : if then . So we can expand
Substitute into and compare coefficients of the linearly independent operators :
So the matrix (zero rows for padding) has orthonormal columns; complete it to an unitary . Viewing each list as a column of kets and padding the eigen-ensemble with zeros, . Likewise , hence
Example: arises from , from (related by ), and from the trine whose Bloch vectors are apart on a great circle. Comparing the trine with needs one zero vector of padding and a unitary.
Let . What is the largest weight with which can appear in an ensemble for ?
Show the answer
0.375
, so and . Bloch-ball check: with , the remainder's Bloch vector stays in the ball iff , i.e. .
Purification
Every mixed state is the shadow of a pure state on a larger system. A pure state purifies if ; is called the reference or purifying system. Two standard constructions:
Fix any eigenbasis of , with for , and define for each purification
These vectors are orthonormal, degenerate eigenvalues included:
Expand in the basis of . Components with vanish, because . This gives Schmidt forms with the same -side vectors:
Define for and extend it isometrically to all of , which is possible because . Then .
Steering: the HJW theorem
Now the punchline. Suppose Bob holds the purifying system of Alice's . Which ensembles can Bob prepare for Alice, just by measuring and announcing nothing?
Take with ; any other purification differs by an isometry on , which Bob can absorb into his measurement. By unitary freedom, for an unitary (padding with for ). Let Bob measure
This is a rank-one POVM on the support of (add as an outcome that never occurs). Outcome projects onto
so it occurs with probability and leaves in . If , extend to orthonormal vectors and sum up to : the become orthonormal, a projective measurement.
So every ensemble decomposition of is physically realized by some measurement on the purifying system. Schrödinger, reacting to the EPR paper, called this steering and found it deeply troubling.
Alice and Bob share . Bob measures on his qubit and finds , i.e. the state . What is Alice's conditional state?
Show the answer
B.
For , , the complex conjugate of Bob's vector. Indeed , so Bob's steers Alice to . For real vectors such as or the conjugation is invisible, which is why and outcomes are perfectly correlated.
Majorization, entropy, and where ensembles matter
(⇒) Unitary freedom relative to the eigen-ensemble gives , and the are orthogonal with squared norms :
The matrix is unistochastic, hence doubly stochastic, so (Hardy–Littlewood–Pólya). (⇐) Conversely, by Horn's theorem every can be written with for a real orthogonal ; then is the desired ensemble.
Shannon entropy is Schur-concave, so every pure-state ensemble pays at least the von Neumann entropy in classical randomness:
Ensembles are invisible to measurements on , yet they are not idle bookkeeping. Quantities defined by optimizing over ensembles, so-called convex roofs, are central to entanglement theory:
Ensembles also give meaning to unravelings of open-system dynamics. The same Lindblad equation can be unraveled into quantum-jump or diffusive trajectories, depending on how the environment, which plays the role of the purifying system, is monitored. Each monitoring scheme picks out an ensemble, and all of them average to the same .
Key takeaways
- Ensembles and give the same iff for a unitary , after zero-padding.
- Purifications exist with and are unique up to an isometry on .
- HJW steering: whoever holds the purification can prepare any ensemble of by measuring alone, without changing , which is precisely why no signal is sent.
- can appear with weight at most ; the possible weight vectors are exactly those majorized by the spectrum, so .
- Optimizations over ensembles (convex roofs such as entanglement of formation) are where ensembles regain operational meaning, and they are hard to compute in general.
Check yourself
In the interactive lesson these are scored with your confidence, and what you miss comes back for review.
Which ensemble has the same density matrix as ?
- The pure state proportional to
Show the answer
C.
The mixture has Bloch vector : length , direction . Its eigen-ensemble consists of the two antipodal states on that axis with weights , which reproduces the Bloch vector .
What is the smallest dimension of a purifying system for a qutrit state with eigenvalues ?
Show the answer
B.
A purification needs , and the Schmidt form achieves it with a single qubit.
Why does the HJW theorem imply that entanglement cannot be used to signal?
- Bob's measurement leaves Alice's conditional states unchanged
- Purifications are unique, so Bob has only one possible measurement
- Steering only works if Bob also sends Alice a quantum system
- Every ensemble Bob can steer Alice into averages to the same , which alone fixes Alice's statistics
Show the answer
D. Every ensemble Bob can steer Alice into averages to the same , which alone fixes Alice's statistics
Bob's choice selects an ensemble , so it does change Alice's conditional states. But Alice does not learn without a classical message, and for every choice Bob can make.
For , what is the largest weight with which a given pure state can appear in an ensemble for ?
Show the answer
A.
, so . It is attained by any orthonormal basis containing , with uniform weights.
Further reading
- L. P. Hughston, R. Jozsa & W. K. Wootters, "A complete classification of quantum ensembles having a given density matrix", Physics Letters A 183, 14 (1993).
- E. Schrödinger, "Probability relations between separated systems", Proceedings of the Cambridge Philosophical Society 32, 446 (1936).
- M. A. Nielsen & I. L. Chuang, Quantum Computation and Quantum Information, Theorem 2.6 and §2.5 (Cambridge, 2010).
- M. A. Nielsen, "Probability distributions consistent with a mixed state", Physical Review A (2000).
- H. M. Wiseman, S. J. Jones & A. C. Doherty, "Steering, entanglement, nonlocality, and the Einstein–Podolsky–Rosen paradox", Physical Review Letters 98, 140402 (2007).