Credits & sources
The work behind
the images.
Research figures supplied by Richard Liu sit alongside new conceptual illustrations of the lab’s planned program.
Multi-objective characterization of a composition library
Research figure supplied by Richard Liu as Fig-2.png, reproduced intact. The figure connects measured responses to Gaussian-process predictions and acquisition maps across an (Al, Sc, B)N library.
Related study: Autonomous Multi-Objective Nanoscale Characterization of Combinatorial (Al, Sc, B)N Films Reveals Composition-Dependent Ferroelectric Regimes.
Multi-objective structure–property learning
Research workflow figure supplied by Richard Liu as fig1.png, reproduced intact. It connects image patches, measured objectives, deep kernel learning, and selection of the next experimental location.
Related preprint: Domain Switching on the Pareto Front: Multi-Objective Deep Kernel Learning in Automated Piezoresponse Force Microscopy.
Genetic search and learned priorities
Method schematic supplied by Richard Liu as fig-TOC.png, reproduced intact. It illustrates candidate generation by a genetic algorithm and prioritization by deep kernel learning. DNA motifs represent candidate encodings.
Conceptual research illustrations
The quantum-materials overview and digital-twin schematic were created with GPT-image for this website, using Richard Liu’s supplied research diagrams as conceptual references. They illustrate research goals and experimental workflows; they contain no measured data.
The digital-twin graphic uses a simplified mechanical surrogate to represent the physics-model component.
Richard Liu portrait and biography
Portrait and biographical source: Northeastern University College of Science, “College of Science welcomes new faculty,” September 23, 2026.
Scientific credit
Research images retain their source authorship and applicable rights. Links above identify related work; the supplied local figures have not been assigned a new blanket license.