Michael Behar
25Patents
3h-index
40Co-inventors
59Inventor score
Filing activity: Dec 28, 2012 → Sep 14, 2023
Most-cited inventions
| Patent | Title | Area | Cited by | Status |
|---|---|---|---|---|
| US10372416B2 | Multiply-accumulate “0” data gating | Physics | 13 | Active |
| US10726583B2 | System and method of encoding and decoding feature maps and weights for a convolutional neural network | Physics | 8 | Active |
| US10467795B2 | Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU | Physics | 4 | Active |
| US10210137B2 | Binary multiplier for binary vector factorization | Physics | 3 | Active |
| US11238338B2 | Real time context dependent deep learning | Emerging Cross-Sectional Technologies | 2 | Active |
| US10853035B2 | Multiply-accumulate “0” data gating | Physics | 2 | Active |
| US11080611B2 | Compression for deep learning in case of sparse values mapped to non-zero value | Emerging Cross-Sectional Technologies | 2 | Active |
| US11231963B2 | Methods and apparatus to enable out-of-order pipelined execution of static mapping of a workload | Physics | 1 | Active |
| US11704564B2 | Real time context dependent deep learning | Emerging Cross-Sectional Technologies | 1 | Active |
| US10606559B2 | Multiply-accumulate “0” data gating | Physics | 1 | Active |
| US11675630B2 | Methods and apparatus to configure heterogenous components in an accelerator | Physics | 1 | Active |
| US12131250B2 | Inner product convolutional neural network accelerator | Physics | 0 | Active |
| US10762685B2 | Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU | Physics | 0 | Active |
| US11847497B2 | Methods and apparatus to enable out-of-order pipelined execution of static mapping of a workload | Physics | 0 | Active |
| US12217101B2 | Methods and apparatus to configure heterogenous components in an accelerator | Physics | 0 | Active |
| US11600035B2 | Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU | Physics | 0 | Active |
| US12223427B2 | Real time context dependent deep learning | Emerging Cross-Sectional Technologies | 0 | Active |
| US11250610B2 | Sub-graph in frequency domain and dynamic selection of convolution implementation on a GPU | Physics | 0 | Active |
| US11656846B2 | Multiply-accumulate “0” data gating | Physics | 0 | Active |
| US12147914B2 | Compression for deep learning in case of sparse values mapped to non-zero value | Emerging Cross-Sectional Technologies | 0 | Active |
| US11763183B2 | Compression for deep learning in case of sparse values mapped to non-zero value | Emerging Cross-Sectional Technologies | 0 | Active |
| US11422939B2 | Shared read—using a request tracker as a temporary read cache | Physics | 0 | Active |
| US9875213B2 | Methods, apparatus, instructions and logic to provide vector packed histogram functionality | Physics | 0 | Active |
| US9189398B2 | Apparatus and method for memory-mapped register caching | Emerging Cross-Sectional Technologies | 0 | Active |
| US11151074B2 | Methods and apparatus to implement multiple inference compute engines | Physics | 0 | Active |
Source: USPTO / EPO open patent data. Inventor disambiguation is heuristic; counts are objective bibliographic measures.