Self-evolving artificial intelligence framework to better decipher short-term large earthquakes.

Large earthquakes (EQs) occur at surprising loci and timing, and their descriptions remain a long-standing enigma. Finding answers by traditional approaches or recently emerging machine learning (ML)-driven approaches is formidably difficult due to data scarcity, interwoven multiple physics, and absent first principles. This paper develops a novel artificial intelligence (AI) framework that can transform raw observational EQ data into ML-friendly new features via basic physics and mathematics and that can self-evolve in a direction to better reproduce short-term large EQs. An advanced reinforcement learning (RL) architecture is placed at the highest level to achieve self-evolution. It incorporates transparent ML models to reproduce magnitude and spatial location of large EQs ([Formula: see text] 6.5) weeks before of the failure. Verifications with 40-year EQs in the western U.S. and comparisons against a popular EQ forecasting method are promising. This work will add a new dimension of AI technologies to large EQ research. The developed AI framework will help establish a new database of all EQs in terms of ML-friendly new features and continue to self-evolve in a direction of better reproducing large EQs.

Tackling Multi-Physics Nano-Scale Phenomena in Capillary Force Lithography with Small Data by Hybrid Intelligence.

The scientific community has been looking for novel approaches to develop nanostructures inspired by nature. However, due to the complicated processes involved, controlling the height of these nanostructures is challenging. Nanoscale capillary force lithography (CFL) is one way to use a photopolymer and alter its properties by exposing it to ultraviolet radiation. Nonetheless, the working mechanism of CFL is not fully understood due to a lack of enough information and first principles. One of these obscure behaviors is the sudden jump phenomenon-the sudden change in the height of the photopolymer depending on the UV exposure time and height of nano-grating (based on experimental data). This paper uses known physical principles alongside artificial intelligence to uncover the unknown physical principles responsible for the sudden jump phenomenon. The results showed promising results in identifying air diffusivity, dynamic viscosity, surface tension, and electric potential as the previously unknown physical principles that collectively explain the sudden jump phenomenon.

Curvature-Adjustable Polymeric Nanolens Fabrication Using UV-Controlled Nanoimprint Lithography.

Nanolenses are gaining importance in nanotechnology, but their challenging fabrication is thwarting their wider adoption. Of particular challenge is facile control of the lens' curvature. In this work, we demonstrate a new nanoimprinting technique capable of realizing polymeric nanolenses in which the nanolens' curvature is optically controlled by the ultraviolet (UV) dose at the pre-curing step. Our results reveal a regime in which the nanolens' height changes linearly with the UV dose. Computational modeling further uncovers that the polymer undergoes highly nonlinear dynamics during the UV-controlled nanoimprinting process. Both the technique and the process model will greatly advance nanoscale science and manufacturing technology.