Electronics and Telecommunications Research Institute

Ju-Yeob Kim

Joo Hyun Lee

Jin Kyu Kim

Mi Young Lee

Byung Jo Kim

Seong Min Kim

Tae-Joong Kim

Min-Hyung Cho

Ki Hyuk Park

Young-deuk Jeon

Seong-Min Kim

Sung Eun Kim

Kwang Il Oh

Seong Mo Park

Jin Ho Han

Jin-Ho Han

Bon-Tae Koo

Byung-Jo Kim

Joo-Hyun Lee

Hye-Ji Kim

A method for processing floating point operations in a multi-processor system including a plurality of single processor cores is provided. In this method, upon receiving a group setting for performing an operation, the plurality of single processor cores are grouped into at least one group according to the group setting, and a single processor core set as a master in the group loads an instruction for performing the operation from an external memory, and performs parallel operations by utilizing floating point units (FUPs) of all single processor cores in the group according to the instructions.

Multi-processor system and method for processing floating point operation thereof

Disclosed herein is an Artificial Intelligence (AI) processor. The AI processor includes multiple NVM AI cores for respectively performing basic unit operations required for a deep-learning operation based on data stored in NVM; SRAM for storing at least some of the results of the basic unit operations; and an AI core for performing an accumulation operation on the results of the basic unit operation.

Artificial intelligence processor and method of processing deep-learning operation using the same

The neuromorphic arithmetic device comprises an input monitoring circuit that outputs a monitoring result by monitoring that first bits of at least one first digit of a plurality of feature data and a plurality of weight data are all zeros, a partial sum data generator that skips an arithmetic operation that generates a first partial sum data corresponding to the first bits of a plurality of partial sum data in response to the monitoring result while performing the arithmetic operation of generating the plurality of partial sum data, based on the plurality of feature data and the plurality of weight data, and a shift adder that generates the first partial sum data with a zero value and result data, based on second partial sum data except for the first partial sum data among the plurality of partial sum data and the first partial sum data generated with the zero value.

Neuromorphic arithmetic device and operating method thereof

An artificial neural network apparatus and an operating method including a plurality of layer processors for performing operations on input data are disclosed. The artificial neural network apparatus may include: a flag layer processor for outputting a flag according to a comparison result between a pooling output value of a current frame and a pooling output value of a previous frame; and a controller for stopping operation of a layer processor which performs operations after the flag layer processor among the plurality of layer processors when the flag is outputted from the flag layer processor, wherein the flag layer processor is a layer processor that performs a pooling operation first among the plurality of layer processors.

Artificial neural network apparatus and operating method for the same

An embodiment of the present invention provides a quantization method for weights of a plurality of batch normalization layers, including: receiving a plurality of previously learned first weights of the plurality of batch normalization layers; obtaining first distribution information of the plurality of first weights; performing a first quantization on the plurality of first weights using the first distribution information to obtain a plurality of second weights; obtaining second distribution information of the plurality of second weights; and performing a second quantization on the plurality of second weights using the second distribution information to obtain a plurality of final weights, and thereby reducing an error that may occur when quantizing the weight of the batch normalization layer.

Quantization method and device for weights of batch normalization layer

Provided is an image recognition device. The image recognition device includes a frame data change detector that sequentially receives a plurality of frame data and detects a difference between two consecutive frame data, an ensemble section controller that sets an ensemble section in the plurality of frame data, based on the detected difference, an image recognizer that sequentially identifies classes respectively corresponding to a plurality of section frame data by applying different neural network classifiers to the plurality of section frame data in the ensemble section, and a recognition result classifier that sequentially identifies ensemble classes respectively corresponding to the plurality of section frame data by combining the classes in the ensemble section.

Image recognition device, operating method of image recognition device, and computing device including image recognition device

A method for controlling a memory from which data is transferred to a neural network processor and an apparatus thereof are provided, the method including: generating prefetch information of data by using a blob descriptor and a reference prediction table after history information is input; reading the data in the memory based on the pre-fetch information and temporarily archiving read data in a prefetch buffer; and accessing next data in the memory based on the prefetch information and temporarily archiving the next data in the prefetch buffer after the data is transferred to the neural network from the prefetch buffer.

Method and apparatus for controlling memory using prefetch information

Provided is a convolutional neural network system including a data selector configured to output an input value corresponding to a position of a sparse weight from among input values of input data on a basis of a sparse index indicating the position of a nonzero value in a sparse weight kernel, and a multiply-accumulate (MAC) computator configured to perform a convolution computation on the input value output from the data selector by using the sparse weight kernel.

Convolutional neural network system and operation method thereof

Provided is an image compression device including an object extracting unit configured to perform convolution neural network (CNN) training and identify an object from an image received externally, a parameter adjusting unit configured to adjust a quantization parameter of a region in which the identified object is included in the image on the basis of the identified object, and an image compression unit configured to compress the image on the basis of the adjusted quantization parameter.

Image compression device

Disclosed herein is a noise cancellation apparatus and method, which select in advance parameters to be used for noise cancellation in a reference voice signal section by generating a reference voice signal in advance before a voice signal is generated, thus improving noise cancellation effects. The noise cancellation apparatus includes a parameter initialization unit for determining an initial value of a parameter to be used for noise cancellation, based on reference signals filtered for respective frequencies, a parameter estimation unit for receiving the initial value of the parameter, and estimating the parameter in response to signals that are input after being filtered for respective frequencies, a gain estimation unit for calculating gains for respective frequencies based on the parameter from the parameter estimation unit, and a gain application unit for cancelling noise by applying the gains to the signals that are input after being filtered for respective frequencies.

Noise cancellation apparatus and method

An apparatus and method for improving voice recognition are disclosed herein. The apparatus includes a standard voice transmission unit, a Mel-frequency cepstrum coefficient (MFCC) generation unit, and an MFCC compensation unit. The standard voice transmission unit generates a standard voice. The MFCC generation unit generates voice feature data (MFCC) based on the utterance of the standard voice before voice recognition. The MFCC compensation unit stores a gain value generated based on the standard voice, and compensates for the distortion of the voice feature data based on the utterance of a user using the gain value during the voice recognition.

Growing community of inventors

Daejeon, South Korea

Ju-Yeob Kim