Abstract: Disclosed is a conjugate gradient acceleration apparatus using band matrix compression in depth fusion technology including a band matrix conversion unit configured to convert an adjacency matrix for correcting depth data acquired from data of an image sensor through deep learning based on depth information acquired from a depth sensor into a band matrix using rows of the adjacency matrix as addresses of query points and columns of the adjacency matrix as the nearest neighbors at the query points, a band matrix compression unit configured to mark an index on each band in order to compress the band matrix and to compress data, a memory unit configured to store tile data of the band matrix, and a band matrix calculation unit configured to perform computation of the band matrix and a transposed band matrix or computation of a symmetric band matrix with respect to the band matrix and a vector.

Abstract: Disclosed herein are an apparatus and method for training a deep neural network. An apparatus for training a deep neural network including N layers, each having multiple neurons, includes an error propagation processing unit configured to, when an error occurs in an N-th layer in response to initiation of training of the deep neural network, determine an error propagation value for an arbitrary layer based on the error occurring in the N-th layer and directly propagate the error propagation value to the arbitrary layer, a weight gradient update processing unit configured to update a forward weight for the arbitrary layer based on a feed-forward value input to the arbitrary layer and the error propagation value in response to the error propagation value, and a feed-forward processing unit configured to, when update of the forward weight is completed, perform a feed-forward operation in the arbitrary layer using the forward weight.

Abstract: An MRAM cell includes a switch unit configured to determine opening and closing thereof by a word line voltage and to activate a current path between a bit line and a bit line bar in an opened state thereof, first and second MTJs having opposite states, respectively, and connected in series between the bit line and the bit line bar, to constitute a storage node, and a sensing line configured to be activated in a reading mode of the MRAM cell, thereby creating data reading information based on a voltage between the first and second MTJs, wherein the first and second MTJs have different ones of a low resistance state and a high resistance state, respectively, in accordance with a voltage drop direction between the bit line and the bit line bar, thereby storing data of 0 or 1.

Abstract: Disclosed is a 3D point cloud-based deep learning neural network acceleration apparatus including a depth image input unit configured to receive a depth image, a depth data storage unit configured to store depth data derived from the depth image, a sampling unit configured to sample the depth image in units of a sampling window having a predetermined first size, a grouping unit configured to generate a grouping window having a predetermined second size and to group inner 3D point data by grouping window, and a convolution computation unit configured to separate point feature data and group feature data, among channel-direction data of 3D point data constituting the depth image, to perform convolution computation with respect to the point feature data and the group feature data, to sum the results of convolution computation by group grouped by the grouping unit, and to derive the final result.

Abstract: Disclosed herein are a floating-point computation apparatus and method using Computing-in-Memory (CIM).

Abstract: Disclosed is an energy-efficient retraining method of a generative neural network for domain-specific optimization, including (a) retraining, by a mobile device, a pretrained generative neural network model with respect to some data of a new user dataset, (b) comparing, by the mobile device, the pretrained generative neural network model and a generative neural network model retrained for each layer with each other in terms of a relative change rate of weights, (c) selecting, by the mobile device, specific layers having high relative change rate of weights, among layers of the pretrained generative neural network model, as layers to be retrained, and (d) performing, by the mobile device, weight update for only the layers selected in step (c), wherein only some of all layers are selected and trained in a retraining process that requires a large amount of operation, whereby rapid retraining is performed in the mobile device.

Abstract: Provided is a deep neural network (DNN) learning accelerating apparatus for deep reinforcement learning, the apparatus including: a DNN operation core configured to perform DNN learning for the deep reinforcement learning; and a weight training unit configured to train a weight parameter to accelerate the DNN learning and transmit it to the DNN operation core, the weight training unit including: a neural network weight memory storing the weight parameter; a neural network pruning unit configured to store a sparse weight pattern generated as a result of performing the weight pruning based on the weight parameter; and a weight prefetcher configured to select/align only pieces of weight data of which values are not zero (0) from the neural network weight memory using the sparse weight pattern and transmit the pieces of weight data of which the values are not zero to the DNN operation core.

Abstract: A method of accelerating training of a low-power, high-performance artificial neural network (ANN) includes (a) performing fine-grained pruning and coarse-grained pruning to generate sparsity in weights by a pruning unit in a convolution core of a cluster in a lower-power, high-performance ANN trainer; (b) selecting and performing dual zero skipping according to input sparsity, output sparsity, and the sparsity of weights by the convolution core, and (c) restricting access to a weight memory during training by allowing a deep neural network (DNN) computation core and a weight pruning core to share weights retrieved from a memory by the convolution core.

Abstract: Disclosed herein are an apparatus and method for training a low-bit-precision deep neural network. The apparatus includes an input unit configured to receive training data to train the deep neural network, and a training unit configured to train the deep neural network using training data, wherein the training unit includes a training module configured to perform training using first precision, a representation form determination module configured to determine a representation form for internal data generated during an operation procedure for the training and determine a position of a decimal point of the internal data so that a permissible overflow bit in a dynamic fixed-point system varies randomly, and a layer-wise precision determination module configured to determine precision of each layer during an operation in each of a feed-forward stage and an error propagation stage and automatically change the precision of a corresponding layer based on the result of determination.

Abstract: A computing device includes a first computing core that generates sparsity data based on a first sign bit and first exponent bits of first data and a second sign bit and second exponent bits based on second data, and a second computing core that outputs a result value of a floating point calculation of the first data and the second data as output data or skips the floating point calculation and outputs the output data having a given value, based on the sparsity data.

Abstract: Disclosed herein are an apparatus and method for training a deep neural network. An apparatus for training a deep neural network including N layers, each having multiple neurons, includes an error propagation processing unit configured to, when an error occurs in an N-th layer in response to initiation of training of the deep neural network, determine an error propagation value for an arbitrary layer based on the error occurring in the N-th layer and directly propagate the error propagation value to the arbitrary layer, a weight gradient update processing unit configured to update a forward weight for the arbitrary layer based on a feed-forward value input to the arbitrary layer and the error propagation value in response to the error propagation value, and a feed-forward processing unit configured to, when update of the forward weight is completed, perform a feed-forward operation in the arbitrary layer using the forward weight.

Abstract: A deep neural network accelerator includes a feature loader that stores input features, a weight memory that stores a weight, and a processing element. The processing element applies 1-bit weight values to the input features to generate results according to the 1-bit weight values, receives a target weight corresponding to the input features from the weight memory, and selects a target result corresponding to the received target weight from among the results to generate output features.

Abstract: A method for providing a user interface through a head mounted display using eye recognition and bio-signals comprises the steps of: (a) moving a cursor to a particular location at which a user gazes by referencing the eye information obtained from a first eyeball that is one of the eyeballs of the user through a camera module when the user gazes at a particular location on an output screen; and (b) supporting in order to provide detailed selection items corresponding to an entity when a certain entity exists in the certain position by referencing the movement information obtained from the eyelid corresponding to a second eyeball that is one of the eyeballs of the user through a bio-signal acquisition module.

Abstract: An iontophoretic drug delivery apparatus comprises an electrode unit comprising a plurality of iontophoresis electrodes and a plurality of tissue resistivity measurement electrodes; a programmable current unit configured to control a current that is supplied to the iontophoresis electrodes to thereby control the amount of drug delivered; an impedance detection unit comprising a detection mode that selectively measures a load resistance value between the iontophoresis electrodes or a tissue resistivity value between the tissue resistivity measurement electrodes so as to monitor the amount of drug delivered; and a control unit configured to control the programmable current unit by determining the amount of drug delivered or whether the drug is to be delivered, based on the load resistance or tissue resistivity value measured in the impedance detection unit.

Abstract: Disclosed is an electrical impedance photographing patch comprising a plurality of electrodes contacting with skin, applied with an electrical signal and arranged on a flexible substrate, wherein the photographing patch measures impedance, i.e., an electrical signal, of the skin of a measurement target placed between the plurality of electrodes and transmits the electrical signal to an external device which restores the electrical signal into a three-dimensional tomographic image and displays the tomographic image.

Abstract: Provided is a near field wireless communication apparatus that uses magnetic coupling and a method for operation of the apparatus, in which the magnetic coupling is used to transmit data or clock information with low power and high efficiency. A pulse generator of the near field wireless communication apparatus generates a pulse signal corresponding to transmission digital data to be transmitted. When the transmission digital data is “1,” the near field wireless communication apparatus modulates the data into the pulse signal and transmits the pulse signal. When the data is “0,” the near field wireless communication apparatus does not output the pulse.

Abstract: Disclosed is an apparatus for continuously and automatically measuring a pulse wave by a non-invasive method to know the state of a cardiovascular system. The apparatus includes an integrated measurement module, a communication power module, and a bio-measurement pad. The integrated measurement module includes an electrocardiogram measurement portion for measuring the electrocardiogram of a subject, a bioelectrical impedance measurement portion for measuring the bioelectrical impedance of the subject by a potential difference, a heart sound measurement portion for measuring the heart sound of the subject, and a controller for measuring and controlling the state of the cardiovascular system of the subject on the basis of a pulse transit time calculated by the electrocardiogram signal measured at the electrocardiogram measurement portion, the bioelectrical impedance signal measured at the bioelectrical impedance measurement portion, and the heart sound signal measured at the heart sound measurement portion.

Abstract: A near field wireless communication apparatus using a magnetic coupling to transmit data or clock information with low power and high efficiency is provided. Thus, when data to be transmitted is “1,” the near field wireless communication apparatus modulates the data into a pulse signal and transmits the pulse signal, and when the data is “0,” the near field wireless communication apparatus does not generate a signal at all.

Abstract: Disclosed are a wearable monitoring apparatus and a driving method thereof. The wearable monitoring apparatus comprises: a sensor unit for measuring a biological signal from a human body, wherein the sensor unit is adhered to a skin; and a control unit for searching a location of the sensor unit, supplying power to the sensor unit, and receiving and processing the biological signal from the sensor unit, wherein the control unit is formed to be wearable.

Abstract: One embodiment of the present invention relates to a sensor for measuring biosignals. The sensor according to said embodiment comprises: a sensor layer formed by stacking a plurality of sensor layers that are attachable to the skin to measure different types of bio signals; a power for supplying power to the sensor layer; and a sensing electrode for sensing biosignals from the human body. The plurality of sensor layers takes the signals sensed by the sensing electrode as an input, and determines whether or not to measure the inputted signals. Then, the relevant sensor layer that can measure the sensed signal is activated.