Here are some papers that caught our eye this month
Solving non-native combinatorial optimization problems using hybrid quantum-classical algorithms
While analog neutral-atom machines are highly performant, it is widely understood that the limitations of problems they can address represent an issue to the technology adoption. In this work, our algorithms team introduces a “non-native hybrid algorithms” paradigm which utilizes the quantum system to provide hints that enhance the performance of classical routine for problems outside the architecture’s native space (i.e. maximum independent sets on unit-disk graphs). Data from Aquila, our 256 quantum computer, demonstrate solutions for Max k-cut problems and more, indeed showcasing the power of this methodology.
Correlated decoding of logical algorithms with transversal gates
Following up, a paper led by Madelyn Cain from Harvard, with participation of QuEra researchers. If you followed up the big news of last December on experiments demonstrating complex circuits with 48 logical neutral atom qubits, you may have read about the advances in error-correction decoding protocols that enabled the work. This paper describes exactly those advances in detail, showcasing how joint decoding of qubits during transversal entangling gates can reduce error propagation and substantially reduce the overheads on rounds of syndrome extraction per gate.
{{Newsletter-signup}}
Compiling Quantum Circuits for Dynamically Field-Programmable Neutral Atoms Array Processors
In the same trend of topics, dynamically reconfiguring atom positions for gate-based quantum computing represents both an opportunity and a challenge. The opportunity is clear, given the recent demonstrations from late last year, but challenge is that the design space for program compilation strategies is incredibly rich and complex. In this work, teams based in UCLA and Harvard put forward a compiler system that demonstrates unequivocally the advantages of reconfigurable atomic connectivity. Reductions in 2-qubit gate counts varying between 1.7x and 5.1x are demonstrated in different applications.
A tweezer array with 6100 highly coherent atomic qubits
Wrapping up the month with experimental advances! Manuel Endres’ group at Caltech showcases experiments where over 6000(!) atoms are loaded intro laser tweezer traps, all while characterizing coherence of single-qubit gates and achieving trap lifetimes of over 20 minutes! The authors emphasize the importance of this result to pave the way for neutral-atom quantum computing with 10,000 qubits in the near future, which corroborates with QuEra’s expected technology roadmap.