The quantum ensemble variational optimization algorithm: Applications to molecular inverse design

Area of Science:

• Computational chemistry
• Quantum computing applications
• Drug discovery

Background:

• Designing molecules with specific properties is complex due to the vast number of possibilities.
• Traditional computational methods struggle with large molecular design spaces (curse of dimensionality).
• Quantum computing presents a promising avenue for tackling these complex computational challenges.

Purpose of the Study:

• Introduce a novel quantum computing method, quantum ensemble variational optimization (QEVO).
• Address the limitations of traditional computational approaches in molecular design.
• Demonstrate QEVO's capability in discovering molecules with specific therapeutic properties.

Main Methods:

• Developed the quantum ensemble variational optimization (QEVO) method for near-term quantum devices.
• Mapped molecular structures to an orthonormal basis of binary strings.
• Utilized a variational ansatz to sample from a superposition state, iteratively optimized for desired properties.
• Employed shallow quantum circuits with a modest number of qubits.

Main Results:

• QEVO successfully navigates combinatorial spaces up to [Formula: see text] solutions.
• Demonstrated potential for designing drug-like molecules with anticancer properties.
• The method operates efficiently on current and early fault-tolerant quantum computing platforms.

Conclusions:

• QEVO provides an efficient quantum approach for molecular design.
• The method can tackle complex problems with significant combinatorial complexity.
• QEVO has broad applicability across various scientific domains requiring optimization in large search spaces.