This inventor holds 1 USPTO granted patent. Top assignees: Constructor Education and Research Genossenschaft, Constructor Technology Ag. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
1 out of 18 
1 out of 30 Years Active: 2026
Title: Ignat Romanov: Machine Learning for Defect Properties in 2D Materials.
Introduction
Ignat Romanov is an inventor associated with Beijing, China. His patent work explores how machine learning can help predict properties of defects in two-dimensional materials.
Latest Patents
Romanov is associated with 1 patent, titled “Sparse Representation for Machine Learning the Properties of Defects in 2D Materials.” The invention describes a method for predicting at least one property of a crystal containing a point defect. It uses sampled material and defect structures to generate data representing the defect, including its coordinates, type, and crystal unit-cell structure.
The method uses a neural network that receives material and ideal crystal unit-cell structures as input and outputs at least one target quantity. It can also receive a cloud of defect points or a global state vector and produce an output vector.
Career Highlights
Romanov is associated with Constructor Technology AG and Constructor Education and Research Genossenschaft. His patent reflects an application of neural networks and sparse representations to the study of material defects.
Collaborations
His listed coworkers include Nikita Kazeev and Andrey Ustyuzhanin.
Conclusion
Ignat Romanov’s patent connects machine learning with materials science by proposing a neural-network method for predicting properties of point defects in two-dimensional materials.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile