Demonstrating an Unconditional Separation Between Quantum and Classical Information Resources
summary
Collaboration between the Simons Institute, Quantinuum and UT Austin implements an unconditional separation: a 12‑qubit trapped‑ion experiment (Quantinuum H1) prepares near‑Haar random states using about 516 entangling gates and reaches roughly 46% fidelity, then runs random Clifford measurements (≈66 entangling gates, ≈96% per‑circuit fidelity). The recorded linear‑cross‑entropy score proves classical replication needs at least 12 qubits vs 62 bits (lower bound; upper bound ≤382), with no hardness assumptions.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:06Hilbert space challenge
- 3:57Bell vs Hilbert tests
- 4:42Unconditional separation goal
- 9:44Why previous separations were not demonstrated
- 12:14Experimental task setup
- 17:42Haar random states with fidelity
- 19:55Haar measurements with Clifford
- 22:11Results and storage bounds
- 23:59Future directions
- 25:46Loopholes and improvements
- 26:59Q&A
related talks
01h 00mQuantum Physics for Software Developers
46mProbing the limits of classical computing with arbitrarily connected quantum circuits
01h 08mPutting Quantum Supremacy on a Rigorous Footing
42mTests for "Quantumness"
56mWhat is Your Logical Qubit?
38mQuantum Technologies: Quantum Computing & Secure Quantum Communication
01h 09mRecent Progress in Quantum Advantage
48mStatistics In Quantum Computational Supremacy