Recogni Inc.

United States of America

Back to Profile

1-42 of 42 for Recogni Inc. Sort by
Query
Aggregations
IP Type
        Patent 31
        Trademark 11
Jurisdiction
        United States 27
        World 8
        Europe 7
Date
New (last 4 weeks) 1
2025 May (MTD) 1
2025 April 2
2025 (YTD) 3
2024 8
See more
IPC Class
G06N 3/04 - Architecture, e.g. interconnection topology 14
G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means 12
G06F 17/15 - Correlation function computation 7
G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising 7
G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation 7
See more
NICE Class
09 - Scientific and electric apparatus and instruments 11
42 - Scientific, technological and industrial services, research and design 2
Status
Pending 2
Registered / In Force 40

1.

Split accumulator with a shared adder

      
Application Number 18679915
Grant Number 12293163
Status In Force
Filing Date 2024-05-31
First Publication Date 2025-05-06
Grant Date 2025-05-06
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Goldman, Gary S.

Abstract

In a multiply accumulate (MAC) unit, an accumulator may be implemented in two or more stages. For example, a first accumulator may accumulate products from the multiplier of the MAC unit, and a second accumulator may periodically accumulate the running total of the first accumulator. Each time the first accumulator's running total is accumulated by the second accumulator, the first accumulator may be initialized to begin a new accumulation period. In one embodiment, the number of values accumulated by the first accumulator within an accumulation period may be a user-adjustable parameter. In one embodiment, the bit width of the input of the second accumulator may be greater than the bit width of the output of the first accumulator. In another embodiment, an adder may be shared between the first and second accumulators, and a multiplexor may switch the accumulation operations between the first and second accumulators.

IPC Classes  ?

  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation
  • G06F 7/509 - AddingSubtracting in bit-parallel fashion, i.e. having a different digit-handling circuit for each denomination for multiple operands, e.g. digital integrators

2.

INFERENCE INDUSTRIES

      
Serial Number 99144187
Status Pending
Filing Date 2025-04-18
Owner Recogni Inc. ()
NICE Classes  ?
  • 09 - Scientific and electric apparatus and instruments
  • 42 - Scientific, technological and industrial services, research and design

Goods & Services

Semiconductors; Integrated circuits; Computer hardware; Data processing equipment; Microprocessors; Microchips Computer software development; Development of computer platforms; Platform as a service (PAAS) featuring computer software platforms for machine learning

3.

INFERENCEWORKS

      
Serial Number 99144189
Status Pending
Filing Date 2025-04-18
Owner Recogni Inc. ()
NICE Classes  ?
  • 09 - Scientific and electric apparatus and instruments
  • 42 - Scientific, technological and industrial services, research and design

Goods & Services

Semiconductors; Integrated circuits; Computer hardware; Data processing equipment; Microprocessors; Microchips Computer software development; Development of computer platforms; Platform as a service (PAAS) featuring computer software platforms for machine learning

4.

Systems and methods for performing matrix multiplication with a plurality of processing elements

      
Application Number 18523615
Grant Number 12045309
Status In Force
Filing Date 2023-11-29
First Publication Date 2024-07-23
Grant Date 2024-07-23
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Goldman, Gary S.

Abstract

In a system with control logic and a processing element array, two modes of operation may be provided. In the first mode of operation, the control logic may configure the system to perform matrix multiplication or 1×1 convolution. In the second mode of operation, the control logic may configure the system to perform 3×3 convolution. The processing element array may include an array of processing elements. Each of the processing elements may be configured to compute the dot product of two vectors in a single clock cycle, and further may accumulate the dot products that are sequentially computed over time.

IPC Classes  ?

  • G06F 17/16 - Matrix or vector computation
  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation

5.

Multiply accumulate (MAC) unit with split accumulator

      
Application Number 18408296
Grant Number 12039290
Status In Force
Filing Date 2024-01-09
First Publication Date 2024-07-16
Grant Date 2024-07-16
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Goldman, Gary S.

Abstract

In a multiply accumulate (MAC) unit, an accumulator may be implemented in two or more stages. For example, a first accumulator may accumulate products from the multiplier of the MAC unit, and a second accumulator may periodically accumulate the running total of the first accumulator. Each time the first accumulator's running total is accumulated by the second accumulator, the first accumulator may be initialized to begin a new accumulation period. In one embodiment, the number of values accumulated by the first accumulator within an accumulation period may be a user-adjustable parameter. In one embodiment, the bit width of the input of the second accumulator may be greater than the bit width of the output of the first accumulator. In another embodiment, an adder may be shared between the first and second accumulators, and a multiplexor may switch the accumulation operations between the first and second accumulators.

IPC Classes  ?

  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation
  • G06F 7/509 - AddingSubtracting in bit-parallel fashion, i.e. having a different digit-handling circuit for each denomination for multiple operands, e.g. digital integrators

6.

Multiply accumulate (MAC) unit with split accumulator

      
Application Number 18408309
Grant Number 12026478
Status In Force
Filing Date 2024-01-09
First Publication Date 2024-07-02
Grant Date 2024-07-02
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Goldman, Gary S.

Abstract

In a multiply accumulate (MAC) unit, an accumulator may be implemented in two or more stages. For example, a first accumulator may accumulate products from the multiplier of the MAC unit, and a second accumulator may periodically accumulate the running total of the first accumulator. Each time the first accumulator's running total is accumulated by the second accumulator, the first accumulator may be initialized to begin a new accumulation period. In one embodiment, the number of values accumulated by the first accumulator within an accumulation period may be a user-adjustable parameter. In one embodiment, the bit width of the input of the second accumulator may be greater than the bit width of the output of the first accumulator. In another embodiment, an adder may be shared between the first and second accumulators, and a multiplexor may switch the accumulation operations between the first and second accumulators.

IPC Classes  ?

7.

Multi-mode architecture for unifying matrix multiplication, 1×1 convolution and 3×3 convolution

      
Application Number 18523632
Grant Number 12007937
Status In Force
Filing Date 2023-11-29
First Publication Date 2024-06-11
Grant Date 2024-06-11
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Goldman, Gary S.

Abstract

In a system with control logic and a processing element array, two modes of operation may be provided. In the first mode of operation, the control logic may configure the system to perform matrix multiplication or 1×1 convolution. In the second mode of operation, the control logic may configure the system to perform 3×3 convolution. The processing element array may include an array of processing elements. Each of the processing elements may be configured to compute the dot product of two vectors in a single clock cycle, and further may accumulate the dot products that are sequentially computed over time.

IPC Classes  ?

  • G06F 15/80 - Architectures of general purpose stored program computers comprising an array of processing units with common control, e.g. single instruction multiple data processors
  • G06F 9/54 - Interprogram communication

8.

Multi-mode architecture for unifying matrix multiplication, 1×1 convolution and 3×3 convolution

      
Application Number 18523627
Grant Number 12008069
Status In Force
Filing Date 2023-11-29
First Publication Date 2024-06-11
Grant Date 2024-06-11
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Goldman, Gary S.

Abstract

In a system with control logic and a processing element array, two modes of operation may be provided. In the first mode of operation, the control logic may configure the system to perform matrix multiplication or 1×1 convolution. In the second mode of operation, the control logic may configure the system to perform 3×3 convolution. The processing element array may include an array of processing elements. Each of the processing elements may be configured to compute the dot product of two vectors in a single clock cycle, and further may accumulate the dot products that are sequentially computed over time.

IPC Classes  ?

  • G06F 17/16 - Matrix or vector computation
  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation

9.

Low power hardware architecture for a convolutional neural network

      
Application Number 18410736
Grant Number 12141685
Status In Force
Filing Date 2024-01-11
First Publication Date 2024-05-02
Grant Date 2024-11-12
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Bodwin, James Michael
  • Joginipally, Pradeep R.
  • Abhiram, Shabarivas
  • Goldman, Gary S.
  • Patz, Martin Stefan
  • Feinberg, Eugene M.
  • Ozceri, Berend

Abstract

Dynamic data quantization may be applied to minimize the power consumption of a system that implements a convolutional neural network (CNN). Under such a quantization scheme, a quantized representation of a 3×3 array of m-bit activation values may include 9 n-bit mantissa values and one exponent shared between the n-bit mantissa values (n

IPC Classes  ?

  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G06F 7/50 - AddingSubtracting
  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation
  • G06N 3/0464 - Convolutional networks [CNN, ConvNet]
  • G06N 3/06 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means

10.

METHODS AND SYSTEMS FOR PROCESSING READ-MODIFY-WRITE REQUESTS

      
Application Number US2023015130
Publication Number 2024/035446
Status In Force
Filing Date 2023-03-13
Publication Date 2024-02-15
Owner RECOGNI INC. (USA)
Inventor
  • Goldman, Gary, S.
  • Radhakrishnan, Ashwin

Abstract

A memory system comprises a plurality of memory sub-systems, each with a memory bank and other circuit components. For each of the memory sub-systems, a first buffer receives and stores a read-modify -write request (with a read address, a write address and a first operand), a second operand is read from the memory bank at the location specified by the read address, a combiner circuit combines the first operand with the second operand, an activation circuit transforms the output of the combiner circuit, and the output of the activation circuit is stored in the memory bank at the location specified by the write address. The first operand and the write address may be stored in a second buffer while the second operand is read from the memory bank. Further, the output of the activation circuit may be first stored in the first buffer before being stored in the memory bank.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 3/06 - Digital input from, or digital output to, record carriers

11.

Methods and systems for processing read-modify-write requests

      
Application Number 18183034
Grant Number 12271624
Status In Force
Filing Date 2023-03-13
First Publication Date 2024-02-15
Grant Date 2025-04-08
Owner Recogni Inc. (USA)
Inventor
  • Goldman, Gary S.
  • Radhakrishnan, Ashwin

Abstract

A memory system comprises a plurality of memory sub-systems, each with a memory bank and other circuit components. For each of the memory sub-systems, a first buffer receives and stores a read-modify-write request (with a read address, a write address and a first operand), a second operand is read from the memory bank at the location specified by the read address, a combiner circuit combines the first operand with the second operand, an activation circuit transforms the output of the combiner circuit, and the output of the activation circuit is stored in the memory bank at the location specified by the write address. The first operand and the write address may be stored in a second buffer while the second operand is read from the memory bank. Further, the output of the activation circuit may be first stored in the first buffer before being stored in the memory bank.

IPC Classes  ?

  • G06F 3/06 - Digital input from, or digital output to, record carriers
  • G11C 11/4093 - Input/output [I/O] data interface arrangements, e.g. data buffers
  • G11C 11/4096 - Input/output [I/O] data management or control circuits, e.g. reading or writing circuits, I/O drivers or bit-line switches
  • H03K 19/173 - Logic circuits, i.e. having at least two inputs acting on one outputInverting circuits using specified components using elementary logic circuits as components

12.

Low power hardware architecture for handling accumulation overflows in a convolution operation

      
Application Number 17806143
Grant Number 12165041
Status In Force
Filing Date 2022-06-09
First Publication Date 2023-12-14
Grant Date 2024-12-10
Owner Recogni Inc. (USA)
Inventor
  • Abhiram, Shabarivas
  • Goldman, Gary S.
  • Huang, Jian Hui
  • Feinberg, Eugene M.

Abstract

In a low power hardware architecture for handling accumulation overflows in a convolver unit, an accumulator of the convolver unit computes a running total by successively summing dot products from a dot product computation module during an accumulation cycle. In response to the running total overflowing the maximum or minimum value of a data storage element, the accumulator transmits an overflow indicator to a controller and sets its output equal to a positive or negative overflow value. In turn, the controller disables the dot product computation module by clock gating, clamping one of its inputs to zero and/or holding its inputs to constant values. At the end of the accumulation cycle, the output of the accumulator is sampled. In response to a clear signal being asserted, the dot product computation module is enabled, and the running total is set to zero for the start of the next accumulation cycle.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 9/30 - Arrangements for executing machine instructions, e.g. instruction decode
  • G06F 17/16 - Matrix or vector computation

13.

Systems for using shifter circuit and 3×3 convolver units to emulate functionality of larger sized convolver units

      
Application Number 17934702
Grant Number 11762946
Status In Force
Filing Date 2022-09-23
First Publication Date 2023-09-19
Grant Date 2023-09-19
Owner Recogni Inc. (USA)
Inventor
  • Goldman, Gary S.
  • Abhiram, Shabarivas

Abstract

Convolution with a 5×5 kernel involves computing the dot product of a 5×5 data block with a 5×5 kernel. Instead of computing this dot product as a single sum of 25 products, the dot product is computed as a sum of four partial sums, where each partial sum is computed as a dot product of a 3×3 data block with a 3×3 kernel. The four partial sums may be computed by a single 3×3 convolver unit over four time periods. During each time period, at least some of the weights received by the 3×3 convolver unit may correspond to a quadrant of weights from the 5×5 kernel. A shifter circuit provides shifted columns (left or right shifted) of the input data to the 3×3 convolver unit, allowing the 3×3 convolver unit access to the 3×3 data block that spatially corresponds to a particular quadrant of weights from the 5×5 kernel.

IPC Classes  ?

  • G06F 17/15 - Correlation function computation
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/16 - Matrix or vector computation

14.

Systems for evaluating a piecewise linear function

      
Application Number 18148673
Grant Number 11645355
Status In Force
Filing Date 2022-12-30
First Publication Date 2023-05-09
Grant Date 2023-05-09
Owner Recogni Inc. (USA)
Inventor
  • Backhus, Gilles J. C. A.
  • Goldman, Gary S.

Abstract

A system for evaluating a piecewise linear function includes a first look-up table with N entries, and a second look-up table with M entries, with M being less than N. Each of the N entries contains parameters that define a corresponding linear segment of the piecewise linear function. The system further includes a controller configured to store a subset of the N entries from the first look-up table in the second look-up table. The system further includes a classifier for receiving an input value and classifying the input value in one of a plurality of segments of a number line. A total number of the segments is equal to M, and the segments are non-overlapping and contiguous. The system further includes a multiplexor for selecting one of the M entries of the second look-up table based on the classification of the input value into one of the plurality of segments.

IPC Classes  ?

  • G06F 17/12 - Simultaneous equations
  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation

15.

Methods and systems for processing read-modify-write requests

      
Application Number 17818876
Grant Number 11630605
Status In Force
Filing Date 2022-08-10
First Publication Date 2023-04-18
Grant Date 2023-04-18
Owner Recogni Inc. (USA)
Inventor
  • Goldman, Gary S.
  • Radhakrishnan, Ashwin

Abstract

A memory system comprises a plurality of memory sub-systems, each with a memory bank and other circuit components. For each of the memory sub-systems, a first buffer receives and stores a read-modify-write request (with a read address, a write address and a first operand), a second operand is read from the memory bank at the location specified by the read address, a combiner circuit combines the first operand with the second operand, an activation circuit transforms the output of the combiner circuit, and the output of the activation circuit is stored in the memory bank at the location specified by the write address. The first operand and the write address may be stored in a second buffer while the second operand is read from the memory bank. Further, the output of the activation circuit may be first stored in the first buffer before being stored in the memory bank.

IPC Classes  ?

  • G06F 3/06 - Digital input from, or digital output to, record carriers

16.

Methods for processing data in an efficient convolutional engine with partitioned columns of convolver units

      
Application Number 17811417
Grant Number 11694069
Status In Force
Filing Date 2022-07-08
First Publication Date 2022-11-03
Grant Date 2023-07-04
Owner Recogni Inc. (USA)
Inventor Feinberg, Eugene M.

Abstract

Contiguous columns of a convolutional engine are partitioned into two or more groups. Each group of columns may be used to process input data. Filter weights assigned to one group may be distinct from filter weights assigned to another group.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/15 - Correlation function computation
  • G06N 5/046 - Forward inferencingProduction systems
  • G06N 3/04 - Architecture, e.g. interconnection topology

17.

Efficient convolutional engine

      
Application Number 17811409
Grant Number 11580372
Status In Force
Filing Date 2022-07-08
First Publication Date 2022-11-03
Grant Date 2023-02-14
Owner Recogni Inc. (USA)
Inventor Feinberg, Eugene M.

Abstract

A hardware architecture for implementing a convolutional neural network.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/15 - Correlation function computation
  • G06N 5/04 - Inference or reasoning models
  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G06N 5/046 - Forward inferencingProduction systems

18.

Efficient convolutional engine

      
Application Number 17811410
Grant Number 11593630
Status In Force
Filing Date 2022-07-08
First Publication Date 2022-11-03
Grant Date 2023-02-28
Owner Recogni Inc. (USA)
Inventor Feinberg, Eugene M.

Abstract

A hardware architecture for implementing a convolutional neural network. Certain ones of the convolver units may be controlled to be active and others may be controlled to be non-active by a controller in order to perform convolution with a striding of greater than or equal to two.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/15 - Correlation function computation
  • G06N 5/04 - Inference or reasoning models
  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G06N 5/046 - Forward inferencingProduction systems

19.

Methods for processing horizontal stripes of data in an efficient convolutional engine

      
Application Number 17811412
Grant Number 11694068
Status In Force
Filing Date 2022-07-08
First Publication Date 2022-11-03
Grant Date 2023-07-04
Owner Recogni Inc. (USA)
Inventor Feinberg, Eugene M.

Abstract

A convolutional engine is configured to process input data that is organized into horizontal stripes. The number of accumulators present in each convolver unit of the convolutional engine may equal a total number of rows of data in each of the horizontal stripes.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/15 - Correlation function computation
  • G06N 5/046 - Forward inferencingProduction systems
  • G06N 3/04 - Architecture, e.g. interconnection topology

20.

Methods for processing vertical stripes of data in an efficient convolutional engine

      
Application Number 17811416
Grant Number 11645504
Status In Force
Filing Date 2022-07-08
First Publication Date 2022-11-03
Grant Date 2023-05-09
Owner Recogni Inc. (USA)
Inventor Feinberg, Eugene M.

Abstract

A convolutional engine is configured to process input data that is organized into vertical stripes.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/15 - Correlation function computation
  • G06N 5/046 - Forward inferencingProduction systems
  • G06N 3/04 - Architecture, e.g. interconnection topology

21.

Low power hardware architecture for a convolutional neural network

      
Application Number 16948164
Grant Number 11915126
Status In Force
Filing Date 2020-09-04
First Publication Date 2022-03-10
Grant Date 2024-02-27
Owner Recogni Inc. (USA)
Inventor
  • Huang, Jian Hui
  • Bodwin, James Michael
  • Joginipally, Pradeep R.
  • Abhiram, Shabarivas
  • Goldman, Gary S.
  • Patz, Martin Stefan
  • Feinberg, Eugene M.
  • Ozceri, Berend

Abstract

Dynamic data quantization may be applied to minimize the power consumption of a system that implements a convolutional neural network (CNN). Under such a quantization scheme, a quantized representation of a 3×3 array of m-bit activation values may include 9 n-bit mantissa values and one exponent shared between the n-bit mantissa values (n

IPC Classes  ?

  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G06N 3/06 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 7/50 - AddingSubtracting
  • G06F 7/544 - Methods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using non-contact-making devices, e.g. tube, solid state deviceMethods or arrangements for performing computations using exclusively denominational number representation, e.g. using binary, ternary, decimal representation using unspecified devices for evaluating functions by calculation
  • G06N 3/0464 - Convolutional networks [CNN, ConvNet]

22.

LOW POWER HARDWARE ARCHITECTURE FOR A CONVOLUTIONAL NEURAL NETWORK

      
Application Number US2021048078
Publication Number 2022/051191
Status In Force
Filing Date 2021-08-27
Publication Date 2022-03-10
Owner RECOGNI INC. (USA)
Inventor
  • Huang, Jian Hui
  • Bodwin, James Michael
  • Joginipally, Pradeep, R.
  • Abhiram, Shabarivas
  • Goldman, Gary, S.
  • Patz, Martin, Stefan
  • Feinberg, Eugene, M.
  • Ozceri, Berend

Abstract

Dynamic data quantization may be applied to minimize the power consumption of a system that implements a convolutional neural network (CNN). Under such a quantization scheme, a quantized representation of a 3x3 array of m-bit activation values may include 9 n- bit mantissa values and one exponent shared between the n-bit mantissa values (n < m); and a quantized representation of a 3x3 kernel with p-bit parameter values may include 9 q-bit mantissa values and one exponent shared between the q-bit mantissa values (q < p). Convolution of the kernel with the activation data may include computing a dot product of the 9 n-bit mantissa values with the 9 q-bit mantissa values, and summing the shared exponents. In a CNN with multiple kernels, multiple computing units (each corresponding to one of the kernels) may receive the quantized representation of the 3x3 array of m-bit activation values from the same quantization-alignment module.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06N 3/04 - Architecture, e.g. interconnection topology

23.

RECOGNI VISION COGNITION MODULE

      
Application Number 018485072
Status Registered
Filing Date 2021-06-04
Registration Date 2021-09-23
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment.

24.

RECOGNI VCM

      
Application Number 018455952
Status Registered
Filing Date 2021-04-19
Registration Date 2021-08-26
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment.

25.

VCM

      
Application Number 018455954
Status Registered
Filing Date 2021-04-19
Registration Date 2021-11-10
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment; all of the aforesaid goods only for use in the field of object recognition for self-driving cars and autonomous and assisted driving of vehicles.

26.

Miscellaneous Design

      
Application Number 018455956
Status Registered
Filing Date 2021-04-19
Registration Date 2021-08-26
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment.

27.

RECOGNI REALTIME OBJECT RECOGNITION

      
Application Number 018455957
Status Registered
Filing Date 2021-04-19
Registration Date 2021-08-26
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment.

28.

RECOGNI

      
Application Number 018455958
Status Registered
Filing Date 2021-04-19
Registration Date 2021-08-26
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment.

29.

RECOGNI

      
Serial Number 90371977
Status Registered
Filing Date 2020-12-10
Registration Date 2023-01-17
Owner Recogni Inc. ()
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment

30.

Cluster compression for compressing weights in neural networks

      
Application Number 16273592
Grant Number 11468316
Status In Force
Filing Date 2019-02-12
First Publication Date 2019-09-19
Grant Date 2022-10-11
Owner Recogni Inc. (USA)
Inventor
  • Backhus, Gilles J. C. A.
  • Feinberg, Eugene M.

Abstract

A method for instantiating a convolutional neural network on a computing system. The convolutional neural network includes a plurality of layers, and instantiating the convolutional neural network includes training the convolutional neural network using a first loss function until a first classification accuracy is reached, clustering a set of F×K kernels of the first layer into a set of C clusters, training the convolutional neural network using a second loss function until a second classification accuracy is reached, creating a dictionary which maps each of a number of centroids to a corresponding centroid identifier, quantizing and compressing F filters of the first layer, storing F quantized and compressed filters of the first layer in a memory of the computing system, storing F biases of the first layer in the memory, and classifying data received by the convolutional neural network.

IPC Classes  ?

31.

Deterministic labeled data generation and artificial intelligence training pipeline

      
Application Number 16273604
Grant Number 10922585
Status In Force
Filing Date 2019-02-12
First Publication Date 2019-09-19
Grant Date 2021-02-16
Owner Recogni Inc. (USA)
Inventor
  • Abhiram, Shabarivas
  • Feinberg, Eugene M.

Abstract

Labeled data is deterministically generated for training or validating machine learning models for image analysis. Approaches are described that allow this training data to be generated, for example, in real-time, and in response to the conditions at the location where images are generated by image sensors.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 3/08 - Learning methods
  • H04N 5/225 - Television cameras
  • G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
  • G06N 3/04 - Architecture, e.g. interconnection topology

32.

CLUSTER COMPRESSION FOR COMPRESSING WEIGHTS IN NEURAL NETWORKS

      
Application Number US2019017781
Publication Number 2019/177731
Status In Force
Filing Date 2019-02-13
Publication Date 2019-09-19
Owner RECOGNI INC. (USA)
Inventor
  • Backhus, Gilles, J.C.A.
  • Feinberg, Eugene, M.

Abstract

FKCFFFF biases of the first layer in the memory, and classifying data received by the convolutional neural network.

IPC Classes  ?

33.

DETERMINISTIC LABELED DATA GENERATION AND ARTIFICIAL INTELLIGENCE TRAINING PIPELINE

      
Application Number US2019017784
Publication Number 2019/177733
Status In Force
Filing Date 2019-02-13
Publication Date 2019-09-19
Owner RECOGNI INC. (USA)
Inventor
  • Abhiram, Shabarivas
  • Feinberg, Eugene, M.

Abstract

Systems, methods, and machine-readable media for deterministically generating labeled data for training or validating machine learning models for image analysis are described. Approaches described herein allow this training data to be generated, for example, in real time, and in response to the conditions at the location where images are generated by image sensors.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 3/04 - Architecture, e.g. interconnection topology
  • G06N 3/08 - Learning methods

34.

SYSTEMS AND METHODS FOR INTER-CAMERA RECOGNITION OF INDIVIDUALS AND THEIR PROPERTIES

      
Application Number US2019017786
Publication Number 2019/177734
Status In Force
Filing Date 2019-02-13
Publication Date 2019-09-19
Owner RECOGNI INC. (USA)
Inventor Abhiram, Shabarivas

Abstract

Systems, methods, and machine-readable media for using a convolutional neural network to generate hash strings corresponding to object instances, and thereby use the characteristic hash strings to recognize the same object instance depicted in images generated at different times and by different camera devices.

IPC Classes  ?

  • G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

35.

Methods for inter-camera recognition of individuals and their properties

      
Application Number 16273609
Grant Number 11429820
Status In Force
Filing Date 2019-02-12
First Publication Date 2019-09-19
Grant Date 2022-08-30
Owner Recogni Inc. (USA)
Inventor Abhiram, Shabarivas

Abstract

A convolutional neural network is used to generate hash strings corresponding to object instances. The characteristic hash strings are used to recognize the same object instance depicted in images generated at different times and by different camera devices.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means
  • G06N 5/04 - Inference or reasoning models
  • G06N 3/08 - Learning methods
  • G06V 20/10 - Terrestrial scenes
  • G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions

36.

Efficient convolutional engine

      
Application Number 16273616
Grant Number 11468302
Status In Force
Filing Date 2019-02-12
First Publication Date 2019-09-19
Grant Date 2022-10-11
Owner Recogni Inc. (USA)
Inventor Feinberg, Eugene M.

Abstract

A hardware architecture for implementing a convolutional neural network.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06F 5/01 - Methods or arrangements for data conversion without changing the order or content of the data handled for shifting, e.g. justifying, scaling, normalising
  • G06F 17/15 - Correlation function computation
  • G06N 5/04 - Inference or reasoning models
  • G06N 3/04 - Architecture, e.g. interconnection topology

37.

Three-dimensional environment modeling based on a multi-camera convolver system

      
Application Number 16273618
Grant Number 10740964
Status In Force
Filing Date 2019-02-12
First Publication Date 2019-09-19
Grant Date 2020-08-11
Owner Recogni Inc. (USA)
Inventor
  • Abhiram, Shabarivas
  • Backhus, Gilles J. C. A.
  • Feinberg, Eugene M.
  • Ozceri, Berend
  • Patz, Martin Stefan

Abstract

A three-dimensional model of the environment of one or more camera devices is determined, in which image processing for inferring the model may be performed at the one or more camera devices.

IPC Classes  ?

  • G06T 17/05 - Geographic models
  • G06N 20/00 - Machine learning
  • H04N 7/18 - Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
  • G06T 7/593 - Depth or shape recovery from multiple images from stereo images
  • G06T 7/246 - Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
  • G06T 7/579 - Depth or shape recovery from multiple images from motion

38.

REAL-TO-SYNTHETIC IMAGE DOMAIN TRANSFER

      
Application Number US2019017782
Publication Number 2019/177732
Status In Force
Filing Date 2019-02-13
Publication Date 2019-09-19
Owner RECOGNI INC. (USA)
Inventor
  • Backhus, Gilles, J.C.A.
  • Abhiram, Shabarivas
  • Feinberg, Eugene, M.

Abstract

Systems, methods, and machine-readable media for deterministically generating labeled data for training or validating machine learning models for image analysis, and for using such machine learning models to determine the contents of real-domain images by using a domain transfer to synthetic-appearing images are described.

IPC Classes  ?

  • G06K 9/62 - Methods or arrangements for recognition using electronic means

39.

EFFICIENT CONVOLUTIONAL ENGINE

      
Application Number US2019017787
Publication Number 2019/177735
Status In Force
Filing Date 2019-02-13
Publication Date 2019-09-19
Owner RECOGNI INC. (USA)
Inventor Feinberg, Eugene, M.

Abstract

A hardware architecture for implementing a convolutional neural network.

IPC Classes  ?

  • G06N 3/063 - Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
  • G06N 3/04 - Architecture, e.g. interconnection topology

40.

THREE-DIMENSIONAL ENVIRONMENT MODELING BASED ON A MULTICAMERA CONVOLVER SYSTEM

      
Application Number US2019017789
Publication Number 2019/177736
Status In Force
Filing Date 2019-02-13
Publication Date 2019-09-19
Owner RECOGNI INC. (USA)
Inventor
  • Abhiram, Shabarivas
  • Backhus, Gilles, J.C.A.
  • Feinberg, Eugene, M.
  • Ozceri, Berend
  • Patz, Martin, Stefan

Abstract

Systems, methods, and machine-readable media for determining a three-dimensional environment model of the environment of one or more camera devices, in which image processing for inferring the model may be performed at the camera devices, are described.

IPC Classes  ?

  • G06T 7/579 - Depth or shape recovery from multiple images from motion
  • G06T 7/246 - Analysis of motion using feature-based methods, e.g. the tracking of corners or segments

41.

RECOGNI

      
Application Number 018098218
Status Registered
Filing Date 2019-07-22
Registration Date 2019-12-04
Owner Recogni Inc. (USA)
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

Semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment.

42.

RECOGNI

      
Serial Number 88286139
Status Registered
Filing Date 2019-02-01
Registration Date 2022-12-13
Owner Recogni Inc. ()
NICE Classes  ? 09 - Scientific and electric apparatus and instruments

Goods & Services

semiconductors; semiconductor chips; computer hardware; recorded computer software featuring artificial intelligence for the operation of computer chips; recorded computer software featuring artificial intelligence for the autonomous driving of vehicles, autonomous navigation, autonomous control of vehicles and assisted driving; recorded computer software featuring artificial intelligence for the collection, compilation, processing, transmission and dissemination of positioning data featuring roadway, geographic, map, route planning, crowd source information, travel information enabling structuring, maintaining and using computerized models of an environment of the vehicle by processing signals from sensors, recognition of landmarks including traffic signs, road profile and lampposts and correcting ego motion estimation; interactive recorded computer software that provides roadway, navigation, geographic, map and travel information; Interactive recorded computer software featuring artificial intelligence for enabling creation or updating of computerized data models of an environment