Systems and methods are disclosed for providing a trusted database system that leverages a small amount of trusted storage to secure a larger amount of untrusted storage. Data are encrypted and validated to prevent unauthorized modification or access. Encryption and hashing are integrated with a low-level data model in which data and meta-data are secured uniformly. Synergies between data validation and log-structured storage are exploited.

The present disclosure provides systems, methods, and computer program products for providing efficient embedding table storage and lookup in machine-learning models. A computer-implemented method may include obtaining an embedding table comprising a plurality of embeddings respectively associated with a corresponding index of the embedding table, compressing each particular embedding of the embedding table individually allowing each respective embedding of the embedding table to be decompressed independent of any other embedding in the embedding table, packing the embedding table comprising individually compressed embeddings with a machine-learning model, receiving an input to use for locating an embedding in the embedding table, determining a lookup value based on the input to search indexes of the embedding table, locating the embedding based on searching the indexes of the embedding table for the determined lookup value, and decompressing the located embedding independent of any other embedding in the embedding table.

An application code hiding apparatus includes a secret code dividing part, a secret code caller generating part, a code analyzing part, a dummy code generating part, a code encrypting part, a code disposing part, a code decryptor generating part, a loader generating part, a decrypted code caller generating part and an unloader generating part. The secret code dividing part divides an application code into a secret code and a normal code. The secret code caller generating part generates a secret code caller. The dummy code generating part generates a dummy code corresponding to the secret code. The code disposing part disposes the dummy code and the encrypted secret code and generates position information thereof. The code decryptor generating part generates a code decryptor. The loader generating part generates a loader. The decrypted code caller generating part generates a decrypted code caller. The unloader generating part generates an unloader.

A method includes defining a lookup table having multiple compressed data values and, for each compressed data value, a range of one or more raw data values associated with the compressed data value. The method also includes compressing raw data using the lookup table. Each raw data value is associated with one of the compressed data values in the lookup table based on a distance ratio involving (i) a distance of the raw data value from a minimum value and (ii) a distance of the raw data value from a maximum value.

The present disclosure provides systems, methods, and computer program products for providing efficient embedding table storage and lookup in machine-learning models. A computer-implemented method may include obtaining an embedding table comprising a plurality of embeddings respectively associated with a corresponding index of the embedding table, compressing each particular embedding of the embedding table individually allowing each respective embedding of the embedding table to be decompressed independent of any other embedding in the embedding table, packing the embedding table comprising individually compressed embeddings with a machine-learning model, receiving an input to use for locating an embedding in the embedding table, determining a lookup value based on the input to search indexes of the embedding table, locating the embedding based on searching the indexes of the embedding table for the determined lookup value, and decompressing the located embedding independent of any other embedding in the embedding table.

A method includes defining a lookup table having multiple compressed data values and, for each compressed data value, a range of one or more raw data values associated with the compressed data value. The method also includes compressing raw data using the lookup table. Each raw data value is associated with one of the compressed data values in the lookup table based on a distance ratio involving (i) a distance of the raw data value from a minimum value and (ii) a distance of the raw data value from a maximum value.

Apparatus, method, and media for permitting use of content. An exemplary method comprises associating a transfer right with content, the transfer right specifying that the content is permitted to be transferred from a first computing device to a second computing device, transferring the content from the first computing device to the second computing device in accordance with the transfer right, updating information associated with the transfer right based on the transfer of the content from the first computing device to the second computing device, and associating a usage right with the content, the usage right corresponding to a utilization of the content, wherein the first computing device includes at least a server mode of operation, and wherein the second computing device includes both a requester mode of operation and a server mode of operation.

An integrated circuit device includes a shuffler, a logic unit and registers each including two or more bit storages. The shuffler receives an address indicating one of the registers and data bits, selects target bit storages at which the data bits are to be stored from among bit storages of the registers depending on a shuffle configuration and the address, stores the data bits into the target bit storages, and transfers the data bits from the target bit storages depending on the shuffle configuration. The logic unit receives the data bits transferred from the shuffler and operates using the received data bits. The shuffle configuration is adjusted when a reset operation is performed.

A system and method for enhancing security for a high security embedded system. The system on chip device including at least one central processing unit (CPU) component, input and output component blocks, an independent hard or soft core dedicated to the input and output blocks, and a built-in, on die interposer, wherein the interposer consists of a field programmable gate array (FPGA) fabric, the FPGA fabric surrounding the components of the system on chip. The method for includes separating system components using a FPGA fabric, redirecting or changing the appearance of system components unknown to other system components, separating system code from security and recovery code, and providing proactive security problem detection and resolutions.

A method for searchable encryption of a system defining a secret key and a public is provided. A data stream cipher can include n elementary data (b.sub.1, b.sub.2, . . . , b.sub.n). The method can include generation of a variate for all elementary data b.sub.j, for values of j from 1 to n, generation of an element function of the public key (g.sup.x(bj),zj) and the variate, the element being associated with a random element of a group of a bilinear environment, the element associated with the random element of the group forming first encryption data (C.sub.j,1). The method can also include generation of a shift factor (g.sup.a.zj−1) function of the variate and the public key, and associated with the random element of the group, the shift factor representing a position of the monomial in the encrypted stream, the shift factor associated with the random element of the group forming second encryption data. The data stream cipher can include the first and second encryption data for all values of j from 1 to n.