Mark Veselov

Title: Mark Veselov: Innovator in Variational Autoencoders

Introduction

Mark Veselov is a prominent inventor based in Moscow, Russia. He has made significant contributions to the field of machine learning, particularly in the development of innovative models that enhance the generation of new objects with specific properties.

Latest Patents

Mark Veselov holds two patents, with his latest invention being a model known as the Variational Autoencoder with a learnable tensor train induced prior (VAE-TTLP). This model is designed to generate new objects, such as molecules, that possess specific properties and biological activity. The VAE-TTLP can be trained to accommodate data that may omit one or more properties of the object, yet still produce an object with the desired characteristics.

Career Highlights

Mark is currently employed at Insilico Medicine IP Limited, where he applies his expertise in machine learning and artificial intelligence. His work focuses on advancing the capabilities of generative models in the context of drug discovery and molecular design.

Collaborations

Some of his notable coworkers include Aleksandr M Aliper and Alexander Zhebrak, who contribute to the innovative research environment at Insilico Medicine.

Conclusion

Mark Veselov's work in the field of variational autoencoders represents a significant advancement in the generation of molecules with tailored properties. His contributions continue to influence the landscape of machine learning and its applications in medicine.