Inventor · disambiguated record
Liangzhen Lai
Also filed as: Lai Liangzhen
15 granted patents·5 pending applications·31 citations·filing 2016–2023
88Inventor score
Top patents by PatentIndex Score
20 records- 0197US11709783B1Tensor data distribution using grid direct-memory access (DMA) controllerMETA PLATFORMS INC·Filed 2020·Granted Jul 25, 2023·14 cites·20 claims
- 0292US12001893B1Distributed synchronization schemeMETA PLATFORMS INC·Filed 2020·Granted Jun 4, 2024·4 cites·20 claims
- 0392US11972349B1Flexible compute array utilization in a tensor processorMETA PLATFORMS INC·Filed 2020·Granted Apr 30, 2024·3 cites·20 claims
- 0492US11704562B1Architecture for virtual instructionsMETA PLATFORMS INC·Filed 2020·Granted Jul 18, 2023·4 cites·20 claims
- 0580US11301545B2Power efficient multiply-accumulate circuitryFACEBOOK TECH LLC·Filed 2019·Granted Apr 12, 2022·3 cites·19 claims
- 0677US11698529B2Systems and methods for distributing a neural network across multiple computing devicesMETA PLATFORMS TECH LLC·Filed 2019·Granted Jul 11, 2023·2 cites·18 claims
- 0775US11954025B2Systems and methods for reading and writing sparse data in a neural network acceleratorMETA PLATFORMS TECH LLC·Filed 2023·Granted Apr 9, 2024·0 cites·20 claims
- 0869US11385864B2Counter based multiply-and-accumulate circuit for neural networkFACEBOOK TECH LLC·Filed 2019·Granted Jul 12, 2022·1 cites·20 claims
- 0968US2022308835A1Counter based multiply-and-accumulate circuit for neural networkMETA PLATFORMS TECH LLC·Filed 2022·Application pending·0 cites
- 1067US2022237262A1Power efficient multiply-accumulate circuitryFACEBOOK TECH LLC·Filed 2022·Application pending·0 cites
- 1158US12265492B2Circular buffer for input and output of tensor computationsMETA PLATFORMS INC·Filed 2023·Granted Apr 1, 2025·0 cites·20 claims
- 1258US2024143525A1Transferring non-contiguous blocks of data using instruction-based direct-memory access (dma)META PLATFORMS INC·Filed 2022·Application pending·0 cites
- 1357US11630770B2Systems and methods for reading and writing sparse data in a neural network acceleratorMETA PLATFORMS TECH LLC·Filed 2019·Granted Apr 18, 2023·0 cites·20 claims
- 1455US12197362B2Batch matrix multiplication operations in a machine learning acceleratorMETA PLATFORMS INC·Filed 2023·Granted Jan 14, 2025·0 cites·20 claims
- 1554US10977002B2System and method for supporting alternate number format for efficient multiplicationFACEBOOK TECH LLC·Filed 2019·Granted Apr 13, 2021·0 cites·20 claims
- 1648US9922152B2Computer implemented system and method for reducing failure in time soft errors of a circuit designADVANCED RISC MACH LTD·Filed 2016·Granted Mar 20, 2018·0 cites·20 claims
- 1745US2021012178A1Systems, methods, and devices for early-exit from convolutionFACEBOOK TECH LLC·Filed 2019·Application pending·0 cites
- 1844US2021012186A1Systems and methods for pipelined parallelism to accelerate distributed processingFACEBOOK TECH LLC·Filed 2019·Application pending·0 cites
- 1941US11675998B2System and method for performing small channel count convolutions in energy-efficient input operand stationary acceleratorMETA PLATFORMS TECH LLC·Filed 2019·Granted Jun 13, 2023·0 cites·18 claims
- 2037US10122384B2Logical interleaverADVANCED RISC MACH LTD·Filed 2016·Granted Nov 6, 2018·0 cites·18 claims
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →