This inventor holds 1 USPTO granted patent and 1 EPO patent. Top assignee: Mitsubishi Electric Research Laboratories, Inc.. Active years: 2026.
Company Filing History:
Years Active: 2026
Title: Aleksei Sholokhov - Innovator in Neural Network Modeling
Introduction
Aleksei Sholokhov is an inventor based in Cambridge, AL (US). He is known for his work in the field of neural networks and their application in controlling high-dimensional physical systems. His innovative approach focuses on utilizing reduced order modeling to enhance the efficiency of systems governed by complex dynamics.
Latest Patent Applications
Aleksei Sholokhov has submitted a patent application titled "Reduced Order Modeling and Control of High Dimensional Physical Systems using Neural Network Model." This application outlines a system and method for training a neural network to control operations of systems characterized by non-linear dynamics represented by partial differential equations (PDEs). The method involves collecting digital representations of time series data that indicate instances of the function space of the system, along with measurements of the system's operational state. Collocation points corresponding to solutions of the PDE are generated, and the neural network is trained using this data to optimize the parameters of a non-linear operator. The architecture of the neural network includes an encoder to transform training data into a latent space, a non-linear operator to propagate these instances, and a decoder to minimize a hybrid loss function.
Conclusion
Aleksei Sholokhov's contributions to the field of neural networks and control systems demonstrate his innovative spirit and commitment to advancing technology. His work has the potential to significantly impact the way complex systems are modeled and controlled.