Abstract: Disclosed is an operation method of a host which controls a computing device performing an artificial intelligence computation. The operation method includes receiving configuration information from the computation device, generating a plurality of lightening weight data by performing lightening on weight data based on the configuration information, generating a plurality of files by performing an aligning operation on the plurality of lightening weight data based on the configuration information, and loading the plurality of files into a memory device of the computing device. The configuration information includes channel information about a plurality of channels between the memory device and an accelerator, which are included in the computing device, and the number of the plurality of files is equal to the number of the plurality of channels.

Abstract: A matrix multiplier includes an input vector scaler configured to generate a scaled input matrix based on a first input matrix and a plurality of scale factors, a first data type converter configured to convert the data type of the scaled input matrix to fixed-point and generate a fixed-point input matrix, a multiplication and accumulation operator array configured to receive the fixed-point input matrix and a plurality of binary vectors, generate a fixed-point output matrix based on the fixed-point input matrix and the plurality of binary vectors, generate the first input matrix and the second input matrix, and generate a first output matrix based on the first input matrix and the second input matrix, and a second data type converter configured to convert the data type of the fixed-point output matrix to a floating point and generate a second output matrix.

Abstract: Disclosed is a storage device, which includes a nonvolatile memory device that stores or reads user data and a controller that controls the nonvolatile memory device, and the nonvolatile memory device includes a memory cell array including a plurality of memory cells that stores data bits corresponding to the user data, a compression circuit that compresses soft-bit data sensed from the plurality of memory cells, and control logic that controls the compression circuit through a plurality of compression stages and transmits a stage control signal to the compression circuit to control whether compression on each of the plurality of compression stages is performed.

Abstract: Disclosed is an accelerator performing an artificial intelligence (AI) computation, which includes a processing element that generates first result data by performing a first computation on first activation data and first weight data loaded from a memory, and a quantizer that generates first output data by performing a quantization on the first result data, and the first activation data, the first weight data, and the first output data are of a low precision type, the first result data is of a high precision type, and the first output data is stored in the memory.

Abstract: Disclosed is a machine learning accelerator which includes a first data controller that stores original length information indicating an original length, receives first data with a first length, and decompresses the first data with the first length to output second data with the original length, a second data controller that stores the original length information, receives third data with a second length shorter than the first length, and decompresses the third data with the second length to output fourth data with the original length, a first accelerator core that receives the second data with the original length from the first data controller and performs a first machine learning-based operation, and a second accelerator core that receives the fourth data with the original length from the second data controller and performs a second machine learning-based operation.

Abstract: A matrix multiplication device including a weight memory circuit, an input matrix buffer, a first processing element array and a second processing element array is provided. The weight memory circuit stores a first pruned sub-matrix and a second pruned sub-matrix. The input matrix buffer receives an input matrix including a plurality of input elements, outputs a first plurality of input elements corresponding to a first plurality of residual weights of the first pruned sub-matrix, and outputs a second plurality of input elements corresponding to a second plurality of residual weights of the second pruned sub-matrix. The first processing element array receives the first plurality of residual weights and the first plurality of input elements, and outputs a first sub-output matrix. The second processing element array receives the second plurality of residual weights and the second plurality of input elements, and output a second sub-output matrix.

Abstract: A matrix multiplier includes an input vector scaler generating a first quantization scaled input vector based on a first input vector, a plurality of common scale coefficients, and first-to-Rth multiplication scale coefficients, a first data type converter generating a first fixed point quantization scaled input vector based on the first quantization scaled input vector, an element array comprising a first processing element generating a first fixed point output element based on the first fixed point quantization scaled input vector and first plurality of quantization sign bits, and a second processing element generating a second fixed point output element based on the first fixed point quantization scaled input vector and second plurality of quantization sign bits, and a second data type converter generating and outputting first and second output elements by converting data types of the first and second fixed point output elements.

Abstract: A matrix multiplier includes an input vector scaler configured to generate a first scaled input vector based on a first input vector and a plurality of quantization scale coefficients, a first data type converter configured to generate a first fixed-point scaled input vector based on the first scaled input vector, a processing element array including a first processing element configured to generate a first fixed-point output element based on the first fixed-point scaled input vector and first plurality of quantization sign values and a second processing element configured to generate a second fixed-point output element based on the first fixed-point scaled input vector and second plurality of quantization sign values, and a second data type converter configured to generate first and second output elements by converting data type of the first and second fixed-point output elements, and to output a first output vector including the first and second output elements.

Abstract: The present disclosure provides method and apparatuses for managing memory of storage system. In some embodiments, a controller of a storage system includes a memory storing a program, and a processor configured to execute the program to determine whether a type of data stored in the memory is at least one of a first data type and a second data type, store, in the memory, a header of the data stored in the memory, based on a first determination that the data stored in the memory is of the first data type, compress the data stored in the memory, based on a second determination that data stored in the memory is of the second data type, and power off the memory based on at least one of the header of the data and the compressed data having been stored in the memory.

Abstract: At least one embodiment provides a computing device including: a controller that receives first input data of a first data type and second input data of a second data type different from the first data type, and outputs a first signal representing the first data type, a second signal representing the second data type, and a clock signal based on the number of bits of the first input data and the second input data, and a computing circuit that performs a multiplication computation the first input data and the second input data based on the first signal, the second signal, and the clock signal and generates output data.

Abstract: Disclosed is an accelerator device which includes an interface circuit that communicates with an external device, a memory that stores first data received through the interface circuit, a polar encoder that performs polar encoding with respect to the first data provided from the memory and to output a result of the polar encoding as second data, and an accelerator core that loads the second data. The first data are compressed weight data, the second data are decompressed weight data, the accelerator core is configured to perform machine learning-based inference based on the second data, and the first data are variable in length.

Abstract: A memory device includes an input unit configured to receive a plain text and output plain blocks and CTS plain block, a multi-core unit including a plurality of encryption/decryption cores configured to encrypt each of the plain blocks provided from the input unit and output cipher blocks in accordance with control of an encryption/decryption core control unit, a CTS core unit including a CTS core configured to encrypt the CTS plain block provided from the input unit into a CTS cipher block, and an output unit configured to receive the cipher blocks and the CTS cipher block and output a cipher text. The CTS plain block is generated through a CTS calculation based on the plain text.

Abstract: Disclosed is an accelerator performing an accumulation operation on a plurality of data, each being a floating point type. A method of operating the accelerator includes loading first data, finding a first exponent, which is a maximum value among exponents of the first data, generating aligned first fractions by performing a bit shift on first fractions of the first data based on the first exponent, and generating a first accumulated value by an accumulation operation on the aligned first fractions, loading second data, finding a second exponent, which is a maximum value among exponents of the second data, and generating a first aligned accumulated value by a bit shift on the first accumulated value, generating aligned second fractions by a bit shift on second fractions of the second data, and generating a second accumulated value by an accumulation operation on the aligned second fractions and the first aligned accumulated value.

Abstract: A memory device includes an input unit configured to receive a plain text and output plain blocks and CTS plain block, a multi-core unit including a plurality of encryption/decryption cores configured to encrypt each of the plain blocks provided from the input unit and output cipher blocks in accordance with control of an encryption/decryption core control unit, a CTS core unit including a CTS core configured to encrypt the CTS plain block provided from the input unit into a CTS cipher block, and an output unit configured to receive the cipher blocks and the CTS cipher block and output a cipher text. The CTS plain block is generated through a CTS calculation based on the plain text.

Abstract: The present disclosure provides method and apparatuses for managing memory of storage system. In some embodiments, a controller of a storage system includes a memory storing a program, and a processor configured to execute the program to determine whether a type of data stored in the memory is at least one of a first data type and a second data type, store, in the memory, a header of the data stored in the memory, based on a first determination that the data stored in the memory is of the first data type, compress the data stored in the memory, based on a second determination that data stored in the memory is of the second data type, and power off the memory based on at least one of the header of the data and the compressed data having been stored in the memory.

Abstract: A method of operating a storage controller includes receiving raw data indicating a series of bits each corresponding to one of threshold voltage states, performing a first state shaping for reducing a number of first target bits of the series of bits, logical values of the first target bits being equal to a logical value of a target threshold voltage state of the threshold voltage states in a first page of plural pages, generating first indicator data that indicates the first target bits based on the first state shaping, compressing the first indicator data, and storing the compressed first indicator data.

Abstract: Disclosed is a method for programming a storage device including a nonvolatile memory device and a storage controller for storing multi-bit data, programming, by the storage controller, the multi-bit data into the nonvolatile memory device based on a pre-programming operation, reading state group data of the multi-bit data generated in the nonvolatile memory device based on a program result of the pre-programming operation, and performing, by the storage controller, error correction decoding on the state group data.

Abstract: A storage device is provided including an interface circuit configured to receive application information from a host; a field programmable gate array (FPGA); a neural processing unit (NPU); and a central processing unit (CPU) configured to select a hardware image from among a plurality of hardware images stored in a memory using the application information, and reconfigure the FPGA using the selected hardware image. The NPU is configured to perform an operation using the reconfigured FPGA.

Abstract: Disclosed is an image sensing device including a plurality of defect detectors each suitable for detecting whether a corresponding target image value is defective, and generating detection information corresponding to a result of the detection, a defect scheduler suitable for sequentially outputting one or more defective image values, which are the target image values detected as defective among the plurality of target image values, based on the detection information, and a defect corrector suitable for correcting the output defective image values.

Abstract: A decoder including a main memory, a flag memory and a decoding logic is provided. The flag memory is configured to store flag data and the decoding logic configured to perform an iteration. Further, the decoding logic is configured to: perform an ith operation using first data, wherein i is a natural number, flag-encode second data that is results obtained by performing the ith operation on the first data, store results obtained by performing the flag encoding on the second data in the flag memory as first flag data if the flag encoding succeeds, and store predetermined second flag data that is different from the first flag data of the second data in the flag memory if the flag encoding fails.