Samsung Electronics Co., Ltd.

Seoul National University

Skhynix Inc.

Snu Rdb Foundation

Sungjoo Yoo

Junhaeng Lee

Eunhyeok Park

Hae Rang Choi

Kiyoung Choi

Youngjae Jin

Junwhan Ahn

Seunghwan Cho

Seungyeop Kang

Provided are a neural network method and an apparatus, the method including obtaining a set of floating point data processed in a layer included in a neural network, determining a weighted entropy based on data values included in the set of floating point data, adjusting quantization levels assigned to the data values based on the weighted entropy, and quantizing the data values included in the set of floating point data in accordance with the adjusted quantization levels.

Neural network method and apparatus

A device includes: a comparator; an exclusive-NOR (XNOR) gate; an accumulator; and a multiplication and accumulation (MAC) operator, wherein a basic block of a neural network comprises a first batch normalization layer, a quantization layer, a convolution layer, an active layer, and a second batch normalization layer, and wherein the basic block is driven by the device by a combination of a first batch normalization operation, a sign function operation, a bitwise convolution operation, an activation function operation, a second batch normalization operation, and a residual connection operation.

Method and device with neural network model

A method and apparatus with neural architecture search are provided. A processor-implemented method includes obtaining target data, sampling a trained first neural network into a plurality of second neural networks, training each of the second neural networks based on a portion of the target data, and selecting a second neural network satisfying a predetermined condition among the trained second neural networks for performing an inference operation.

Apparatus and method for neural architecture searching with target data

A neural processor and a control method of the neural processor are provided. The neural processor includes plurality of processing element groups, wherein each of the processing element groups includes a plurality of processing elements configured to perform a vector operation, an overflow accumulator configured to be engaged by a processing element in which an overflow or underflow occurs from among the plurality of processing elements, and a register configured to store information indicating the processing element as an owner processing element.

Neural processor and control method of neural processor

A semiconductor device includes a cell circuit including a plurality of memory arrays, and a control circuit configured to control the cell circuit. A memory array of the plurality of memory arrays has a plurality of sub-arrays including a first sub-array and a second sub array, and an array connecting circuit configured to connect bit lines of the first sub-array to respective corresponding bit lines of the second sub-array according to a copy signal. The semiconductor device may further include a partial sum circuit configured to perform charge sharing between a plurality of bit lines of the first sub-array.

Semiconductor device performing in-memory processing and operation method thereof

A method for pruning a neural network includes initializing a plurality of threshold values respectively corresponding to a plurality of layers included in the neural network; selecting one of the plurality of layers; adjusting the threshold value of the selected layer; and adjusting a plurality of weights respectively corresponding to a plurality of synapses included in the neural network.

Method and device for pruning a neural network

A computing apparatus may process an operation. The computing apparatus may output information regarding an aggregation operation and an operand corresponding to a variable stored in a memory, store information regarding an operator and the aggregation operands regarding the aggregation operation, perform a first partial operation with respect to the aggregation operands and store a result value of the first partial operation, and process the aggregation operation based on storing the variable, performing a second partial operation with respect to the result value of the first partial operation stored in the cache and the operand corresponding to the variable, and storing a result value of the second partial operation.

Growing community of inventors

Seoul, South Korea

Sungjoo Yoo