Harnessing Atom Loss for Below-Threshold Quantum Computing with Neutral Atoms

Atom loss is often viewed as a major challenge for neutral atom quantum computers. In this talk, I will present our recent theoretical and experimental advances that instead leverage atom loss as a resource, enabling the first below-threshold quantum computation demonstration with neutral atoms.

First, I will introduce our theoretical framework and delayed-erasure decoder, which corrects loss errors even without precise location information. Compatible with all error-correcting codes and logical circuits, this approach enables structure-aware decoding and teleportation-based computation. Simulations show improved logical performance with increasing loss fraction in many important subroutines.

Second, I will present our recent experiment using arrays of up to 448 neutral atoms to implement all core components of a scalable QEC architecture. These include multi-round surface code correction with below-threshold performance, lattice surgery, and universal logic via 3D codes. Atom loss detection and mid-circuit qubit re-use enable deep logical circuits with hundreds of teleportations.

Together, these results show how atom loss can be transformed from a limitation into a tool for advancing fault-tolerant quantum computing.

References:

Theory paper:  https://arxiv.org/abs/2502.20558; Experimental paper: https://arxiv.org/abs/2506.20661

Gefen Baranes holds a B.Sc. in Electrical Engineering and Physics from the Technion – Israel Institute of Technology. She is currently a Ph.D. student at MIT and Harvard, working under the supervision of Mikhail Lukin, Vladan Vuletić, and Susanne Yelin. Her research focuses on developing scalable quantum technologies, including quantum computers and quantum networks. She combines theoretical design with experimental implementation, with particular emphasis on neutral atom arrays and multi-node entanglement using solid-state defects.