This inventor holds 5 USPTO granted patents. Top assignees: Progress, Inc., Other. Active years: 2016-2022.
Location History:
- Darmstadt, DE (2016 - 2019)
- Quedlinburg, DE (2022)
Company Filing History:
Years Active: 2016-2022
Title: Dmitri Pescianschi: Innovator in Neural Networks and Matrix Processing
Introduction
Dmitri Pescianschi is a notable inventor based in Darmstadt, Germany. He has made significant contributions to the fields of neural networks and matrix processing, holding a total of five patents. His innovative work focuses on enhancing the efficiency and functionality of neural networks and memory devices.
Latest Patents
One of Dmitri's latest patents is titled "Neural network and method of neural network training." This invention describes a neural network that includes inputs for receiving signals, synapses with corrective weights, and neurons that generate a neuron sum. The controller in this network receives a desired output signal and modifies the corrective weights to minimize deviation from this signal. Additionally, a structure-forming module rearranges connections during training, while a signal allocation module distributes input signals among network elements in real-time. Another significant patent is for a "Memory device and matrix processing unit utilizing the memory device." This apparatus features a three-dimensional slice access memory that allows simultaneous read/write access to an entire data matrix, enhancing data processing capabilities.
Career Highlights
Dmitri has worked with various companies throughout his career, including Progress, Inc. His experience in these organizations has contributed to his expertise in developing advanced technologies in neural networks and matrix processing.
Collaborations
Dmitri has collaborated with notable individuals in his field, including Boris Zlotin and Vladimir Proseanic. These partnerships have likely fostered innovation and the exchange of ideas, further enhancing his contributions to technology.
Conclusion
Dmitri Pescianschi is a prominent inventor whose work in neural networks and matrix processing has the potential to influence various technological advancements. His patents reflect a commitment to innovation and efficiency in data processing.