Harvard theorists find much simpler circuits for port-based teleportation

Started by Penguin79, Today at 02:43 AM

Previous topic - Next topic

0 Members and 2 Guests are viewing this topic.

Topic: Harvard theorists find much simpler circuits for port-based teleportation   Views(Read 66 times)
Active members in this topic:
Penguin79(1)

Penguin79

Two Harvard theorists have found a far simpler way to build circuits for an unusual form of quantum teleportation. Quantum Zeitgeist reports on a preprint by Anurag Anshu and Quynh T. Nguyen titled A simple proof of unitary parallelization via port-based teleportation

In port based teleportation, the sender and receiver share many entangled pairs, and the receiver ends up with the teleported state on one of N ports. Unlike standard teleportation, no correcting operation is needed at the receiving end, which makes it useful in theory for certain computing tasks

Previous circuits for this had depth that grew polynomially with the number of ports and the dimension of the system. The new construction reaches polylogarithmic depth and avoids heavy mathematical tools such as mixed Schur transforms

The paper also offers a simpler proof that any multi qubit operation can be parallelised to polynomial depth, by preparing special states called Choi states. In effect, applying an operation becomes a matter of preparing a state. The article says this independently confirms recent work by other researchers including a team at OpenAI

This is pure theory, and it does not count the cost of error correction or of preparing huge numbers of high quality entangled pairs. Even so, simplifying a complex construction often opens the door to practical uses. Elegant proofs tend to get built on quickly

Could port based teleportation ever be practical? And how important are simpler proofs for moving theory towards hardware?

Trained so hard the GPU asked for a break