Abstract: An embodiment of an integrated circuit may comprise a hardware compressor to compress data, the hardware compressor including circuitry to store input data in a history buffer, compute one or more code tables based on the input data, and compute a compression stream header based on the computed one or more code tables. Other embodiments are disclosed and claimed.

Abstract: A lossless data compressor prevents normalization overruns on-the-fly as symbol occurrence counts are rounded to generate symbol frequencies, allowing an encoding table generator to generate encoding table entries without waiting for the symbol frequency table to finish filling. Rounding errors are accumulated as symbols are normalized and compensated for by reducing a symbol frequency when the symbol frequency is at least 2 and the accumulated errors have exceeded a threshold. The symbol frequency is also reduced when the number of remaining states in the encoding table is insufficient for a number of remaining unprocessed symbols and states for a current encoding table entry. Since error compensation occurs as symbols are being normalized, encoding table generation is not forced to wait for all symbols in the block to be processed, reducing latency. Three pipeline stages can operate on three input blocks: symbol counting, normalization/error compensation/encoding table generation, and data encoding.

Abstract: Disclosed herein relates to a method that improves the accuracy of producing family trees. The DNA of a target individual is processed to find a matching individual. Using the known family tree of the matching individual, multiple candidate family trees are generated with multiple proposed placements for the target individual. For each candidate family tree, a genetic likelihood for a proposed relationship and the other DNA test takers in the family tree. A birth-year probability is determined by identifying a most recent common ancestor (MRCA). The birth-year probability is based on the number of years between the target individual and the matching individual and a normal distribution of ages for parent-child age differences in a population. The genetic likelihood is converted to a genetic probability so that it can be compared with or added to the birth-year probability. Based on the two probabilities, the candidate family trees are sorted.

Abstract: In a first method, a wireless device estimates a delay profile of a channel impulse response, CIR, for a channel between a network node and the wireless device, compresses the delay profile using a compression function, and transmits the compressed delay profile. The compression function includes a first function and a quantizer. The first function is configured to receive input data and reduce a dimension of the input data. In a second method, a network node receives a compressed delay profile of CIR for a channel between a network node and a wireless device, decompresses the compressed delay profile using a decompression function, and estimates a position of the wireless device based on at least the decompressed delay profile. The decompression function includes a first function which is configured to receive input data and provide output data in a higher dimensional space than the input data.

Abstract: A multi-task (MTL) process is adapted to the single-task learning (STL) case, i.e., when only a single task is available for training. The process is formalized as pseudo-task augmentation (PTA), in which a single task has multiple distinct decoders projecting the output of the shared structure to task predictions. By training the shared structure to solve the same problem in multiple ways, PTA simulates the effect of training towards distinct but closely-related tasks drawn from the same universe. Training dynamics with multiple pseudo-tasks strictly subsumes training with just one, and a class of algorithms is introduced for controlling pseudo-tasks in practice.

Abstract: Methods and apparatus relating to an Application Programming Interface (API) for fine grained low latency decompression within a processor core are described. In an embodiment, a decompression Application Programming Interface (API) receives an input handle to a data object. The data object includes compressed data and metadata. Decompression Engine (DE) circuitry decompresses the compressed data to generate uncompressed data. The DE circuitry decompress the compressed data in response to invocation of a decompression instruction by the decompression API. The metadata comprises a first operand to indicate a location of the compressed data, a second operand to indicate a size of the compressed data, a third operand to indicate a location to which decompressed data by the DE circuitry is to be stored, and a fourth operand to indicate a size of the decompressed data. Other embodiments are also disclosed and claimed.

Abstract: According to one embodiment, a memory system includes a shift register memory and a controller. The shift register memory includes data storing shift strings. The controller changes a shift pulse, which is to be applied to the data storing shift strings from which first data is read by applying a first shift pulse, to a second shift pulse to write second data to the data storing shift strings and to read the second data from the data storing shift strings. The controller creates likelihood information of data read from the data storing shift strings in accordance with a read result of the second data. The controller performs soft decision decoding for the first data using the likelihood information.

Abstract: Techniques are presented to stitch existing virtual private networks (VPNs), such as MPLS based VPNs, with virtual private clouds (VPCs) in public cloud data centers. The stitching architecture can be realized by configuring a virtual routing application (VRA) in the VPCs and configuring virtual routing applications and a virtual routing application controller in the existing VPN. For VPCs in public clouds that do not have a VRA, traffic can be default routed to VPCs with a VRA.

Abstract: A method and apparatus that facilitates lossless data compression employing seed bits and recursive processing. The compression steps operate on the initial binary representation of the input digital data stream by representing the digital data stream using two data sets namely a rule data set and a seed data set, wherein the rules data is a dictionary, and the seed data is a unique binary string defined in the course of the recursive processing having a size that is less than the original data string.

Abstract: A battery pack includes a cell module assembly, a primary power output, and a secondary output connector. The cell module assembly includes several lithium-ion battery cells connected in parallel. The primary power output includes a pair of terminals for connection to equipment to be powered by the battery pack. The secondary output connector includes a secondary power output for connection to equipment to be powered by the battery pack and includes an enable input. The primary power output is configured to supply electrical power at a voltage higher than the secondary power output connector. The secondary power output is controllable through the enable input to selectively supply electrical power from the secondary power output.

Abstract: Methods and apparatus are disclosed to model consumer choices.

Abstract: A signal processing device includes an acquisition unit that acquires identification information associated with data obtained by decoding a reception signal that includes a signal included in a first layer and a signal included in a different layer and superimposed on the signal of the first layer in a power axis direction, the data being output to a post-stage processing unit, a decoding unit that decodes the signal of the different layer on the basis of a result of decoding of the signal of the first layer after the decoding of the signal of the first layer, and an output control unit that outputs data obtained by decoding the signal of the different layer to the post-stage processing unit without outputting data obtained by decoding the signal of the first layer to the post-stage processing unit, in a case where data identified by the identification information is data obtained by decoding the signal of the different layer.

Abstract: An apparatus includes a processor to: receive an indication of ability of a node device to provide a resource for executing application routines, at least one identifier of at least one image including an executable routine stored within a cache of the node device, and an indication of at least one revision level of the at least one image; analyze the ability to provide the resource; in response to being able to support execution of the application routine, identify a first image in a repository; compare identifiers to determine whether there is a second image including a matching executable routine; in response to a match, compare revision levels; and in response to the revision level of the most recent version of the first image being more recent, retrieve the most recent version of the first image from the repository, and store it within the node device.

Abstract: Methods, systems, and computer programs for compressing nucleic acid sequence data. A method can include obtaining nucleic acid sequence data representing: (i) a read sequence, and (ii) a plurality of quality scores, determining whether the read sequence includes at least one “N” base, based on a determination that the read sequence includes at least one “N” base, generating, by one or more computers, a first encoding data set by using a first encoding process to encode each set of four quality scores of the read sequence into a single byte of memory, and using a second encoding process to encode the first encoded data set, thereby compressing the data to be compressed.

Abstract: A disclosed information handling system includes an edge device communicatively coupled to a cloud computing resource. The edge device is configured to respond to receiving, from an internet of things (IoT) unit, a numeric value for a parameter of interest by determining a compressed encoding for the numeric value in accordance with a non-lossless compression algorithm. The edge device transmits the compressed encoding of the numeric value to the cloud computing resource. The cloud computing resource includes a decoder communicatively coupled to the encoder and configured to respond to receiving the compressed encoding by generating a surrogate for the numeric value. The surrogate may be generated in accordance with a probability distribution applicable to the parameter of interest. The compression algorithm may be a clustering algorithm such as a k-means clustering algorithm.

Abstract: Methods, systems, and apparatus, including computer-readable storage media for hardware compression and decompression. A system can include a decompressor device coupled to a memory device and a processor. The decompressor device can be configured to receive, from the memory device, compressed data that has been compressed using an entropy encoding, process the compressed data using the entropy encoding to generate uncompressed data, and send the uncompressed data to the processor. The system can also include a compressor device configured to generate, from uncompressed data, a probability distribution of codewords, generate a code table from the probability distribution, and compress incoming data using the generated code table.

Abstract: Disclosed is a neural network structure enabling efficient training of the network and a method thereto. The structure is a ladder-type structure wherein one or more lateral input(s) is/are taken to decoding functions. By minimizing one or more cost function(s) belonging to the structure the neural network structure may be trained in an efficient way.

Abstract: On-chip spread spectrum synchronization between spread spectrum sources is provided. A spread spectrum amplitude of a signal of a spread spectrum reference clock is obtained using one or more delay lines of one or more delay elements in a skitter circuit. A spread width of the spread spectrum amplitude of the signal is determined, using one or more sticky latches in the skitter circuit, based on one or more edges of the signal. A delay line of the one or more delay elements corresponding to a falling edge of the spread width of the signal is identified using combinational circuitry of the skitter circuit. A spread spectrum signal of a spread spectrum slave clock is synchronized with the signal of the spread spectrum reference clock based on the delay line.

Abstract: A method of data compression in which the total size of the compressed data is determined and based on that determination, the bit depth of the input data may be reduced before the data is compressed. The bit depth that is used may be determined by comparing the calculated total size to one or more pre-defined threshold values to generate a mapping parameter. The mapping parameter is then input to a remapping element that is arranged to perform the conversion of the input data and then output the converted data to a data compression element. The value of the mapping parameter may be encoded into the compressed data so that it can be extracted and used when subsequently decompressing the data.

Abstract: There is provided a decoding circuit including; a first decoding unit that decodes a first signal from a multiplexed signal in which the first signal and a second signal are multiplexed in an LDM (Layered Division Multiplexing) system; and a second decoding unit that decodes the second signal from the multiplexed signal using the decoding result of the decoded first signal, wherein the second signal is selectively decoded based on noise information related to a reception state of the multiplexed signal.

Abstract: Devices, methods, and systems for encoding data as DNA are provided. An encoder device can include circuitry to encode a data file having a bit sequence encoding data and to generate a virtual DNA (VDNA) sequence of virtual nucleotide bases (Vnb) that reversibly encodes the bit sequence of the data file, divide the VDNA sequence into a plurality of VDNA fragments, associate each VDNA fragment with an archive library sequence (Arc_SEQ), and generate a read instruction (READ) sequence of differences between each VDNA fragment and each associated Arc_SEQ including sufficient instruction to facilitate regeneration of each VDNA fragment from each associated Arc_SEQ. A codeword sequence (Code_SEQ) is additionally generated for each VDNA fragment that includes a codename identifying the associated Arc_SEQ, the READ sequence associated with the VDNA fragment, and an index sequence (Idx_SEQ) including an index mapping of the VDNA fragment in the VDNA sequence.

Abstract: An encoder encodes transform coefficients of a transform coefficient block representing a block of a picture using a scan pattern which sequentially traverses the transform coefficients of the transform block by encoding absolute values of quantization levels of the transform coefficients. The absolute values are binarized with a binarization comprising a first part of the binarization below a cutoff value and a second part of the binarization, prefixed by a codeword of the first part of the binarization for the cutoff value, above the cutoff value. The encoder adaptively, based on previously encoded transform coefficients, decreases the cutoff value to zero monotonically during encoding of the transform coefficients so that the binarization comprises the second part of the binarization and not the first part of the binarization.

Abstract: A method of reducing transmission power for an encoded data stream includes the steps of receiving an incoming data stream having equal probability for a plurality of incoming data bits, assigning a symbol scheme to the received data bits of the incoming data stream according to probabilities of occurrence of individual ones of the received data bits, and transmitting an outgoing data stream according to the assigned symbol scheme having a second average transmit power, different than the first average transmit power, for a plurality of outgoing symbols.

Abstract: According to one embodiment, a memory system includes a first compression unit, a second compression unit, a non-volatile memory, a first decoding unit, a conversion unit and an output unit. The first compression unit is configured to output second data obtained by compressing first data. The second compression unit is configured to output third data obtained by compressing the second data. Fourth data based on the third data is written to the non-volatile memory. The first decoding unit is configured to decode the third data based on the fourth data to the second data. The conversion unit is configured to acquire fifth data by converting a format of the second data. The output unit is configured to output the fifth data to a host.

Abstract: A storage system includes a storage device and a host device. The storage device includes a nonvolatile memory device having a first size and a first volatile memory device having a second size smaller than the first size and configured to operate as a cache memory with respect to the nonvolatile memory device. The first volatile memory device is configured to allow a first bus portion access to cache data stored in the first volatile memory device. The host device is configured to generate a cache table corresponding to information in the cache data stored in the first volatile memory device and configured to read the cache data stored in the first volatile memory device via the first bus portion based on the cache table.

Abstract: Techniques and systems are provided for implementing a convolutional neural network. One or more convolution accelerators are provided that each include a feature line buffer memory, a kernel buffer memory, and a plurality of multiply-accumulate (MAC) circuits arranged to multiply and accumulate data. In a first operational mode the convolutional accelerator stores feature data in the feature line buffer memory and stores kernel data in the kernel data buffer memory. In a second mode of operation, the convolutional accelerator stores kernel decompression tables in the feature line buffer memory.

Abstract: Provided are computing systems and methods directed to active learning and may provide advantages or improvements to active learning applications for skewed data sets. A challenge in training and developing high-quality models for many supervised learning scenarios is obtaining labeled training examples. Provided are systems and methods for active learning on a training dataset that includes both labeled and unlabeled datapoints. In particular, the systems and methods described herein can select (e.g., at each of a number of iterations) a number of the unlabeled datapoints for which labels should be obtained to gain additional labeled datapoints on which to train a machine-learned model (e.g., machine-learned classifier model). Generally, provided are cost-effective methods and systems for selecting data to improve machine-learned models in applications such as the identification of content items in text, images, and/or audio.

Abstract: A waveform shaping circuit includes a capacitor, an impedance element, a switch circuit, and a switch control circuit. Opposite ends of the capacitor are connected to the input and output, respectively. One end of the impedance element supplies a target constant voltage to the output end of the capacitor. The switch circuit has a switch without a diode. When on, the switch applies the target constant voltage to the output. When off, it does not. The switch control circuit switches the switch on during a high voltage period in an AC component of an input pulse signal and switches the switch off during a low voltage period of the AC component. The circuit shapes the input pulse signal into an output pulse signal whose peak-to-peak voltage is equivalent to a peak-to-peak voltage of the AC component and whose voltage during the high voltage period is at the target constant voltage.

Abstract: A system and method to process audio data received over the ARC or eARC interface of HDMI from audio sources are provided. A media device may receive compressed audio data in a number of data formats. The media device may convert between the audio formats provided by the audio sources and the audio formats supported by audio playback devices. The media device may inspect frames of audio data to determine if the frames are to be decoded. The frame may be decoded and subsequently encoded into the data formats supported by the audio playback devices. To reduce latency, the media device may enable a pass-through mode to bypass the decoding of the frames to allow the frames to be decoded at the audio playback devices. A bi-directional loopback application may route audio data received over the ARC or eARC interface from the audio sources to the audio playback devices.

Abstract: A semiconductor device has a timer unit and a processing unit. The timer unit includes a binary counter, a first converter that converts a first count value output from the binary counter to a gray code to output as first gray code data. The processing unit includes a first synchronizer that captures the first gray code data transferred from the timer unit in synchronization with the system clock signal and outputs the captured first gray code data as second gray code data, and a fault detection unit that generates a data for fault detection based on the first gray code data transferred from the timer unit and compares a second count value based on the second gray code data with a third counter value based on the data for fault detection.

Abstract: In deep learning, and in particular, for data compression for allreduce in deep learning, a gradient may be compressed for synchronization in a data parallel deep neural network training for allreduce by sharing a consensus vector between each node in a plurality of nodes to ensure identical indexing in each of the plurality of nodes prior to performing sparse encoding.

Abstract: An apparatus for downmixing a multi-channel signal having at least two channels, has: a downmixer for calculating a downmix signal from the multi-channel signal, wherein the downmixer is configured to calculate the downmix using an absolute phase compensation, so that a channel having a lower energy among the at least two channels is only rotated or is rotated stronger than a channel having a greater energy in calculating the downmix signal; and an output interface for generating an output signal, the output signal having information on the downmix signal.

Abstract: Performance of an encoding process and a decoding process for a sound signal is enhanced. A representative value calculating part 110 calculates, for each frequency section by a plurality of samples fewer than the number of frequency samples of a sample sequence of a frequency domain signal corresponding to an input acoustic signal, from the sample sequence of the frequency domain signal, a representative value of the frequency section from sample values of samples included in the frequency section, for each of predetermined time sections. A signal companding part 120 obtains, for each of the predetermined time sections, a frequency domain sample sequence obtained by multiplying a weight according to a function value of the representative value by a companding function for which an inverse function can be defined and each of the samples corresponding to the representative value in the sample sequence of the frequency domain signal, as a sample sequence of a weighted frequency domain signal.

Abstract: A computer-implemented method for improving compression of predictive models includes generating an unlabeled simulated data set by expanding an initial data set, and generating a labeled data set by predicting the unlabeled, simulated data set using a complex model to output a plurality of labels. The method also includes training a relatively simple neural network using the labeled data set.

Abstract: This disclosure generally relates to system design for a handheld ultrasound imaging system, which may generate an ultrasound image of tissue. The handheld ultrasound imaging system may comprise: an ultrasound transducer array to generate ultrasound waves comprising a wavelength; a processor coupled to the ultrasound transducer array, the processor configured with instructions that when executed cause the processor to: receive ultrasound data derived from the ultrasound transducer array and generate the plurality of ultrasound images, the plurality of ultrasound images comprising an axial resolution capable of resolving the tissue structure no more than the wavelength, and output the plurality of ultrasound images.

Abstract: A method for generating a bitstream indicative of an object based audio program is described. The bitstream comprises a sequence of containers. A first container of the sequence of containers comprises a plurality of substream entities for a plurality of substreams of the object based audio program and a presentation section. The method comprises determining a set of object channels. The method further comprises providing a set of object related metadata for the set of object channels. In addition, the method comprises inserting a first set of object channel frames and a first set of object related metadata frames into a respective set of substream entities of the first container. Furthermore, the method comprises inserting presentation data into the presentation section.

Abstract: A compression scheme can be selected for an input data stream based on characteristics of the input data stream. For example, when the input data stream is searched for pattern matches, input stream characteristics used to select a compression scheme can include one or more of: type and size of an input stream, a length of a pattern, a distance from a start of where the pattern is to be inserted to the beginning of where the pattern occurred previously, a gap between two pattern matches (including different or same patterns), standard deviation of a length of a pattern, standard deviation of a distance from a start of where the pattern is to be inserted to the beginning of where the pattern occurred previously, or standard deviation of a gap between two pattern matches. Criteria can be established whereby one or more characteristics are used to select a particular encoding scheme.

Abstract: A video decoding method and a video decoding device are provided, and the video decoding method includes the following steps. A to-be-decoded bin string is received. A plurality of first current syntax elements in the to-be-decoded bin string are decoded based on a first context model until a decoding output value of the last one of the first current syntax elements matches a specific bin pattern. When the first current syntax elements are decoded, in response to that a state machine is switched from a first decoding operation mode to a second decoding operation mode, a second current syntax element that exists based on the last one of the first current syntax elements is decoded.

Abstract: An encrypted search uses hash vectorization (HV) models, which are secure, one-way hash indices that are produced by a compression process, such as a modeling phase of Lempel-Ziv (LZ) parsing. Each HV model includes a hash filter and a chain vector. The hash filter is a Boolean quotient filter similar to Bloom filters, and allows for the quick elimination of negative query assessments. The hash filter is followed by a chain vector, which provides spatial modeling of hashed elements throughout the compressed and encrypted data, thereby providing increased levels of accuracy, efficiency, and query expressiveness as compared with known techniques.

Abstract: Various approaches for accelerating data access to a computer memory and predicate evaluation includes storing, in the computer memory, (i) base data as multiple base columns, (ii) multiple sketched columns each corresponding to a base column in the base data and having smaller code values compared thereto, and (iii) a compression map for mapping one or more base columns to the corresponding sketched column; applying the compression map to a query having a predicate; determining data on the sketched column that satisfies the predicate; and evaluating the predicate based at least in part on the determined data on the sketched column without accessing the base column in the base data.

Abstract: The present disclosure provides a computation device. The computation device is configured to perform a machine learning computation, and includes an operation unit, a controller unit, and a conversion unit. The storage unit is configured to obtain input data and a computation instruction. The controller unit is configured to extract and parse the computation instruction from the storage unit to obtain one or more operation instructions, and to send the one or more operation instructions and the input data to the operation unit. The operation unit is configured to perform operations on the input data according to one or more operation instructions to obtain a computation result of the computation instruction. In the examples of the present disclosure, the input data involved in machine learning computations is represented by fixed-point data, thereby improving the processing speed and efficiency of training operations.

Abstract: An uplink multiple input-multiple output (MIMO) transmitter apparatus includes a transmitter chain that includes a sigma-delta circuit that creates a summed (sigma) signal and a difference (delta) signal from two original signals to be transmitted. These new sigma and delta signals are amplified by power amplifiers to a desired output level before having two signals reconstructed from the amplified sigma and amplified delta signals by a second circuit. These reconstructed signals match the two original signals in content but are at a desired amplified level relative to the two original signals. The reconstructed signals are then transmitted through respective antennas as uplink signals. By employing this uplink MIMO transmitter apparatus, it is possible to use smaller power amplifiers, which may reduce footprint, power consumption, and costs of the uplink MIMO transmitter apparatus.

Abstract: A visual localization support system is provided. The visual localization support system includes one or more guidance indicators place on a road surface of a roadway, wherein the one or more guidance indicators each include a matrix barcode that uniquely identifies a location by latitude, longitude, and altitude, and describes an affine shape of the guidance indicator.

Abstract: Systems and methods for improving the performance and stability of bonding radios are provided. One method includes receiving a packet from a client device. Next, the method includes determining whether the received packet is an expected next packet and transmitting the received packet to a next destination if the received packet is the expected next packet. In an event the received packet is not the expected next packet, transmitting the received packet to a queue, setting a timer to wait for the expected next packet, and transmitting a message to the sender of the received packet requesting that the expected next packet be sent.

Abstract: Entropy agnostic data encoding includes: receiving, by an encoder, input data including a bit string; generating a plurality of candidate codewords, including encoding the input data bit string with a plurality of binary vectors, wherein the plurality of binary vectors includes a set of deterministic biased binary vectors and a set of random binary vectors; selecting, in dependence upon a predefined criteria, one of the plurality of candidate codewords; and transmitting the selected candidate codeword to a decoder.

Abstract: The present disclosure relates to a system and method for compressing a dataset. The dataset can be divided in to a plurality of groups. Each group can be converted independently into corresponding text file using dictionary coding technique. The corresponding text files can be compressed independently into corresponding compressed files. Finally, all the corresponding compressed files can be combined together to generate a complete compressed data.

Abstract: A scalable encoding architecture is able to support multiple encoder versions without requiring new versions of the architecture to be generated and supported for each encoder version. An encoding profile template can be generated that includes the available settings for an encoder. A versioned schema can be generated that defines the settings for a particular encoder version. The schema can be persisted such that when a new channel or media stream is to be generated, the schema can be loaded and its values validated in order to cause the encoder to encode the media content using the appropriate values. An interface presented to the user to generate the template or channel can be dynamically generated such that only the applicable and/or relevant settings for a particular encoder version are displayed.

Abstract: The present disclosure relates to electronic device, communication method, decoding method, and medium. There is provided an electronic device comprising processing circuitry. The processing circuitry is configured to determine a codec capability of at least one of a first communication apparatus and a second communication apparatus, wherein determining the codec capability comprises determining whether the codec capability comprises a capability of performing decoding process based on quantum computing. The processing circuitry is further configured to determine codec information for communication between the first communication apparatus and the second communication apparatus based on the codec capability.

Abstract: An encoding apparatus includes a memory and a processor configured to acquire text data, specify a first dynamic dictionary among a plurality of dynamic dictionaries based on attribute information of a first word included in the text data, register the first word in association with a first dynamic code in the first dynamic dictionary, and encode the first word into the first dynamic code.

Abstract: A sensor data are compressed on field devices using a representation is provided. The field device immediately decompresses the compressed data in order to detect a deviation. If there is a deviation, then a cloud storage receives the sensor data as raw uncompressed data. A cloud component receives a trigger signal from the field device, indicating that the representation used by the field device for compression does not sufficiently describe the sensor data. The cloud component then learns a new representation by retrieving and analyzing all data stored in the cloud storage. The method and field device provide robust, compression-based data acquisition. They improve quality and precision of the data captured by the field devices. As the representation in the field device can be updated, it becomes possible to accommodate changes in the device setup. The cloud infrastructure provides automatic learning of the representation in the cloud.