Thomas Boesch
34Patents
5h-index
22Co-inventors
69Inventor score
Filing activity: Feb 16, 2001 → Oct 17, 2023
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US6772267B2 | Multi-portal bridge for providing network connectivity | Electricity | 90 | Expired |
| US10402527B2 | Reconfigurable interconnect | Physics | 12 | Active |
| US10417364B2 | Tool to create a reconfigurable interconnect framework | Physics | 9 | Active |
| US11227086B2 | Reconfigurable interconnect | Physics | 9 | Active |
| US11531873B2 | Convolution acceleration with embedded vector decompression | Electricity | 5 | Active |
| US11586907B2 | Arithmetic unit for deep learning acceleration | Physics | 5 | Active |
| US11610362B2 | Data volume sculptor for deep learning acceleration | Physics | 3 | Active |
| US12073308B2 | Hardware accelerator engine | Physics | 1 | Active |
| US11687762B2 | Acceleration unit for a deep learning engine | Physics | 1 | Active |
| US11675943B2 | Tool to create a reconfigurable interconnect framework | Physics | 1 | Active |
| US11562115B2 | Configurable accelerator framework including a stream switch having a plurality of unidirectional stream links | Physics | 1 | Active |
| US11593609B2 | Vector quantization decoding hardware unit for real-time dynamic decompression for parameters of neural networks | Physics | 1 | Active |
| US11836608B2 | Convolution acceleration with embedded vector decompression | Electricity | 0 | Active |
| US10872186B2 | Tool to create a reconfigurable interconnect framework | Physics | 0 | Active |
| US11710032B2 | Pooling unit for deep learning acceleration | Emerging Cross-Sectional Technologies | 0 | Active |
| US10977854B2 | Data volume sculptor for deep learning acceleration | Physics | 0 | Active |
| US11823771B2 | Streaming access memory device, system and method | Physics | 0 | Active |
| US11360667B2 | Tagged memory operated at lower vmin in error tolerant system | Emerging Cross-Sectional Technologies | 0 | Active |
| US11442700B2 | Hardware accelerator method, system and device | Electricity | 0 | Active |
| US12386506B2 | Tagged memory operated at lower VMIN in error tolerant system | Emerging Cross-Sectional Technologies | 0 | Active |
| US11977971B2 | Data volume sculptor for deep learning acceleration | Physics | 0 | Active |
| US10726177B2 | Reconfigurable interconnect | Physics | 0 | Active |
| US12118451B2 | Deep convolutional network heterogeneous architecture | Physics | 0 | Active |
| US11880759B2 | Vector quantization decoding hardware unit for real-time dynamic decompression for parameters of neural networks | Physics | 0 | Active |
| US11740870B2 | Convolutional network hardware accelerator device, system and method | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.