Technologies for an accelerator interface over Ethernet are disclosed. In the illustrative embodiment, a network interface controller of a compute device may receive a data packet. If the network interface controller determines that the data packet should be pre-processed (e.g., decrypted) with a remote accelerator device, the network interface controller may encapsulate the data packet in an encapsulating network packet and send the encapsulating network packet to a remote accelerator device on a remote compute device. The remote accelerator device may pre-process the data packet (e.g., decrypt the data packet) and send it back to the network interface controller. The network interface controller may then send the pre-processed packet to a processor of the compute device.

Verification of discriminative models includes receiving an input; receiving a prediction from a discriminative model for the input; encoding, using an encoder, a latent variable based on the input; decoding, using a decoder, a reconstructed input based on the prediction and the latent variable; and determining, using an anomaly detection module, whether the prediction is reliable based on the input, the reconstructed input, and the latent variable. The encoder and the decoder are jointly trained to maximize an evidence lower bound of the encoder and the decoder. In some embodiments, the encoder and the decoder are further trained using a disentanglement constraint between the prediction and the latent variable. In some embodiments, the encoder and the decoder are further trained without using inputs that are out of a distribution of inputs used to train the discriminative model or that are adversarial to the discriminative model.

A computer-implemented method, computer program product and computing system for: processing an unencoded data file to identify a plurality of file segments; mapping each of the plurality of file segments to a portion of a dictionary file to generate a plurality of mappings, wherein each of the plurality of mappings includes a starting location and a length, thus generating a related encoded data file based, at least in part, upon the plurality of mappings; and storing the related encoded data file on a cloud-based storage platform.

A computer-implemented method, computer program product and computing system for: processing an unencoded data file to identify a plurality of file segments, wherein the unencoded data file is a dataset for use with an EHR process; mapping each of the plurality of file segments to a portion of a dictionary file to generate a plurality of mappings that each include a starting location and a length, thus generating a related encoded data file based, at least in part, upon the plurality of mappings; receiving a request to manipulate the unencoded data file from the EHR process; and processing the related encoded data file based, at least in part, upon the plurality of mappings and the dictionary file to generate a modified encoded data file that represents the requested manipulations of the unencoded data file.

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.

Methods and devices for encoding a point cloud. A current node associated with a sub-volume is split into further sub-volumes, each further sub-volume corresponding to a child node of the current node, and, at the encoder, an occupancy pattern is determined for the current node based on occupancy status of the child nodes. A probability distribution is selected from among a plurality of probability distributions based on occupancy data for a plurality of nodes neighbouring the current node. The encoder entropy encodes the occupancy pattern based on the selected probability distribution to produce encoded data for the bitstream and updates the selected probability distribution. The decoder makes the same selection based on occupancy data for neighbouring nodes and entropy decodes the bitstream to reconstruct the occupancy pattern.

A deep neural network (“DNN”) module compresses and decompresses neuron-generated activation data to reduce the utilization of memory bus bandwidth. The compression unit receives an uncompressed chunk of data generated by a neuron in the DNN module. The compression unit generates a mask portion and a data portion of a compressed output chunk. The mask portion encodes the presence and location of the zero and non-zero bytes in the uncompressed chunk of data. The data portion stores truncated non-zero bytes from the uncompressed chunk of data. A decompression unit receives a compressed chunk of data from memory in the DNN processor or memory of an application host. The decompression unit decompresses the compressed chunk of data using the mask portion and the data portion.

A method for loop escape analysis includes receiving a set of executable computer instructions stored on a storage medium, and determining a number of inputs to a loop associated with a data structure, storage space that would be saved by compressing the data structure, and a size of new elements required to compress the data structure. Upon reaching an end of the loop, the method determines whether to compress the data structure based on a comparison between the size of the new elements and the saved storage space. In response to determining to compress the data structure, the method compresses the data structure.

Lossy methods and hardware for compressing data and the corresponding decompression methods and hardware are described. The lossy compression method comprises dividing a block of pixels into a number of sub-blocks and then analysing, for each sub-block, and selecting one of a candidate set of lossy compression modes. The analysis may, for example, be based on the alpha values for the pixels in the sub-block. In various examples, the candidate set of lossy compression modes comprises at least one mode that uses a fixed alpha channel value for all pixels in the sub-block and one or more modes that encode a variable alpha channel value.

A method for text compression comprises recognizing a prefix string of one or more text characters preceding a target string of a plurality of text characters to be compressed. The prefix string is provided to a natural language generation (NLG) model configured to output one or more predicted continuations each having an associated rank. If the one or more predicted continuations include a matching predicted continuation relative to the next one or more text characters of the target string, the next one or more text characters are compressed as an NLG-type compressed representation. If no predicted continuations match the next one or more text characters of the target string, a longest matching entry in a compression dictionary is identified. The next one or more text characters of the target string are compressed as a dictionary-type compressed representation that includes the dictionary index value of the longest matching entry.