Precise atom manipulation through deep reinforcement learning

Abstract

Atomic-scale manipulation in scanning tunneling microscopy has enabled the creation of quantum states of matter based on artificial structures and extreme miniaturization of computational circuitry based on individual atoms. The ability to autonomously arrange atomic structures with precision will enable the scaling up of nanoscale fabrication and expand the range of artificial structures hosting exotic quantum states. However, the \textit{a priori} unknown manipulation parameters, the possibility of spontaneous tip apex changes, and the difficulty of modeling tip-atom interactions make it challenging to select manipulation parameters that can achieve atomic precision throughout extended operations. Here we use deep reinforcement learning (DRL) to control the real-world atom manipulation process. Several state-of-the-art reinforcement learning techniques are used jointly to boost data efficiency. The reinforcement learning agent learns to manipulate Ag adatoms on Ag(111) surfaces with optimal precision and is integrated with path planning algorithms to complete an autonomous atomic assembly system. The results demonstrate that state-of-the-art deep reinforcement learning can offer effective solutions to real-world challenges in nanofabrication and powerful approaches to increasingly complex scientific experiments at the atomic scale.

References in corpus (23)

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
• Atomic-scale control of graphene magnetism using hydrogen atoms
• Topological states in engineered atomic lattices
• Design and characterization of electronic fractals
• Atom Transfer and Single-Adatom Contacts
• Strain-Dependence of Surface Diffusion: Ag on Ag(111) and Pt(111)
• A kilobyte rewritable atomic memory
• Designer quantum states of matter created atom-by-atom
• Binary Atomic Silicon Logic
• Site determination and thermally assisted tunneling in homogenous nucleation
• Autonomous robotic nanofabrication with reinforcement learning
• Two-Electron Spin Correlations in Precision Placed Donors in Silicon
• Quantum spins and hybridization in artificially-constructed chains of magnetic adatoms on a superconductor
• p Orbital flat band and Dirac cone in the electronic honeycomb lattice
• An atomic Boltzmann machine capable of on-chip learning
• Boosting Soft Actor-Critic: Emphasizing Recent Experience without Forgetting the Past
• Deep reinforcement learning-designed radiofrequency waveform in MRI
• Single-Atom Gating of Quantum State Superpositions
• Deep Reinforcement Learning for Efficient Measurement of Quantum Devices
• Kondo Effect and Surface-State Electrons
• Quantum confinement of Dirac quasiparticles in graphene patterned with subnanometer precision
• Screening and atomic-scale engineering of the potential at a topological insulator surface
• Exploring Electron Beam Induced Atomic Assembly via Reinforcement Learning in a Molecular Dynamics Environment

Cited by in corpus (6)

• Precise large-scale chemical transformations on surfaces: deep learning meets scanning probe microscopy with interpretability
• Machine Learning for Estimation and Control of Quantum Systems
• Hamiltonian inference from dynamical excitations in confined quantum magnets
• Autonomous convergence of STM control parameters using Bayesian Optimization
• Integration of Scanning Probe Microscope with High-Performance Computing: fixed-policy and reward-driven workflows implementation
• Mic-hackathon 2024: Hackathon on Machine Learning for Electron and Scanning Probe Microscopy