This inventor holds 5 USPTO granted patents and 2 published patent applications. Top assignees: 3D Fusion, Inc., Rembrandt 3D Holding Ltd, Other. Active years: 2013-2022.
Location History:
- New York, NY (US) (2013 - 2017)
- Naples, FL (US) (2022)
Company Filing History:
1 out of 1 
1 out of 1 
3 out of 834,453 Years Active: 2013-2022
Title: Ilya Sorokin: Innovator in Neural Networks and Memory Devices
Introduction
Ilya Sorokin is a prominent inventor based in New York, NY (US), known for his contributions to the fields of neural networks and memory devices. With a total of 5 patents, Sorokin has made significant strides in enhancing the intelligence and learning capabilities of neural networks.
Latest Patents
One of Sorokin's latest patents is titled "System and method of training a neural network." This invention presents a modified model of the Rosenblatt perceptron, which significantly increases both the intelligence and learning speed of the perceptron and other arbitrary neural network topologies. The method incorporates a modified error correction technique and an improved 'Back propagation' algorithm. Additionally, it introduces a cost of connection at the synapses to enhance the manageability of the perceptron topology. The learning process is optimized by transitioning from nested recursive iterations to successive learning for different layers, and by adjusting the learning speed parameter to allow for immediate error compensation.
Another notable patent is "Memory device and matrix processing unit utilizing the memory device." This invention features a matrix processing apparatus equipped with a three-dimensional slice access memory and an input/output block. The slice access memory is organized into cell slices, each capable of storing an entire selected data matrix, allowing for simultaneous read/write access to the entire data matrix.
Career Highlights
Throughout his career, Ilya Sorokin has worked with various companies, including 3D Fusion, Inc. His innovative work has positioned him as a key figure in the development of advanced technologies in neural networks and memory processing.
Collaborations
Sorokin has collaborated with notable individuals in the field, including Stephen Blumenthal and Edmund Mark Hooper. Their combined expertise has contributed to the advancement of Sorokin's inventions.
Conclusion
Ilya Sorokin's contributions to neural networks and memory devices highlight his innovative spirit and dedication to advancing technology. His patents reflect a deep understanding of complex systems and a commitment to improving the efficiency of learning processes.