Abstract: To generate information appropriate for a receiver of the information, a federated learning apparatus includes: a training section which trains a first prediction model that predicts an evaluation value corresponding to a combination of a user and an evaluation target on which the evaluation value is not obtained, using a first training data set including (i) evaluation values of users on evaluation targets and (ii) attribute values of the evaluation targets; a parameter information transmitting section which transmits, to a server apparatus, first parameter information indicating the first prediction model; a parameter information obtaining section which obtains, from the server apparatus, integrated parameter information obtained by integrating the first parameter information and second parameter information indicating a second prediction model trained using a second training data set; and an updating section which updates the first prediction model by replacing the first parameter information with the inte

Abstract: A server device includes an acquisition unit that acquires, from each of a plurality of client devices, information representing a value of a node constituting a decision tree, the information being determined based on the learning data held by the own device of each client device, and a determination unit that determines the value of the node constituting the decision tree by integrating the acquired results. The decision tree is learned by determination of the value of each node constituting the decision tree by the determination unit.

Abstract: A learning device includes an acquisition unit that acquires a local model corresponding to a feature value held by the own device, a residual calculation unit that calculates a difference between an output of a vertical federated learning model having been learned previously and an output of the local model acquired by the acquisition unit, and an additional tree learning unit that learns an additional tree to be added to the local model acquired by the acquisition unit, on the basis of the result of calculation by the residual calculation unit and the feature value held by the own device.

Abstract: Provided is a technique to prevent an attribute inference attack against a federated learning model. An information processing apparatus (2) includes: a detection section (21) that detects, as an outlier, an output value which falls outside a normal range in terms of magnitude of a value among output values of a decision tree which is undergoing federated learning; and a determination section (22) that determines, in a case where the outlier has been detected, that the decision tree has vulnerability to an attack to infer an attribute value included in training data which has been used to train the decision tree.

Abstract: A processing device includes an acquisition unit and a specifying unit. The acquisition unit acquires, from a decision tree that is a learned model and includes a plurality of nodes, score information representing a value according to the number of pieces of data that fell to each of the nodes, among a plurality of pieces of training data used for training of the decision tree. The specifying unit specifies a possible range that the value of an unknown feature may take, on the basis of the score information acquired by the acquisition unit. The unknown feature is a part of the features included in the training data.

Abstract: A machine learning apparatus according to the embodiment including: n (n is an integer greater than or equal to 2) inference units which are machine learning models trained using training data; and a classifier configured to classify an input data and to output an output data. A first inference unit from among the n inference units performs inference based on the input data when the output data of the classifier is a first value. At least one inference unit other than the first inference unit is trained using the input data when the output data of the classifier is the first value as the training data.

Abstract: A machine learning method including a first learning phase for training parameters ? of a learning model f by performing machine learning using a first dataset as a training data with a correct answer label; and a second learning phase for training parameters ? of a defender u and parameters ? of identifier h by performing machine learning using member data contained in the first dataset and non-member data contained in a second dataset. The second learning phase alternately performs, a first step for updating the parameters ? of the identifier h using the identification result when the first input result and the second input result are input to the identifier h; and a second step for updating the parameters ? of the defender u using the first output result, the second output result and the identification result.

Abstract: An inference apparatus according to the embodiment including: an inference unit which is machine learning model trained using training data and configured to inference based on input data; a determination unit configured to determine whether input data is the training data or not based on inference data indicating a result of inference when the input data is input to the inference unit; and an output unit configured to output the inference data as output data when the determination unit determines that the input data is not the training data, and configured to output an alternative data different from the inference data as the output data when the determination unit determines that input data is the training data.

Abstract: The invention includes a first learning phase that a machine learning is performed using first dataset to create a learning model f; and a second learning phase that the first or a second dataset is randomly selected; a result by inputting the first and the second datasets to the learning model f is inputted to a discriminator h having a parameter ?; a machine learning is performed using a result and a ground truth data to train the parameter ?, the result being obtained by having the discriminator h discriminate whether the input data belongs to the first or the second dataset; and when the first dataset is selected, a result by inputting the data of the first dataset to the learning model f is inputted to a defender u to train a parameter ? thereof by using an output of the discriminator h through the defender u.

Abstract: The terminal apparatus comprises a machine learning part that can execute a process of computing a first model update parameter of a first neural network using training data and a process of computing a second model update parameter of a second neural network using training data for a simulated attack; an encryption processing part that encrypts the first, the second model update parameter using a predetermined homomorphic encryption; a data transmission part that transmits the encrypted first, second model update parameters to a predetermined computation apparatus; and an update part that receives from the computation apparatus model update parameters of the first, the second neural networks computed using the first, the second model update parameters received from another terminal apparatus and updates the first, the second neural networks.

Abstract: A decoding apparatus includes a plurality-of-bits decoding part configured to receive an input vector obtained by adding a message encrypted by a trapdoor function and an error vector including an element(s) conforming with a discrete Gaussian distribution, and decode a plurality of bits from a lower bit of the message based on the input vector in correctness with a predetermined probability; and a confirmation calculation part configured to determine in parallel whether the decoded plurality of bits are correct or not, wherein the message is encrypted by taking an inner product with a vector including a power of two as an element(s).

Abstract: A system includes a plurality of nodes, an individual one of which transmits data to which a group signature is attached, and a plurality of management servers that are directly connected to each other. An individual one of the plurality of management servers includes a ledger for managing data received from the nodes. Addition of data to the ledger of at least one of the plurality of management servers is reflected on the ledgers of the other management servers.

Abstract: An authentication system is provided with: a user device; user side assistance device(s) to assist user authentication that authenticates a user of the user device, and apparatus authentication that authenticates the user device; and an apparatus authentication server device to perform apparatus authentication in association with the user device. The user side assistance device(s) use distributed shares of verification information to perform multi-party computation for user authentication in association with the user device, and use distributed shares of a secret key generated by the user device, to perform multi-party computation for apparatus authentication in association with the user device.

Abstract: A system includes a modification apparatus and an execution apparatus. The modification apparatus modifies a program including a conditional branching statement(s) such that, after first processing corresponding to execution processing in the conditional branching statement(s) is executed, whether to reflect a result(s) of the first processing on a variable(s) used in the program as a processing result(s) of the execution processing is determined based on a conditional expression(s) in the conditional branching statement(s). The execution apparatus executes the modified program in an execution environment protected at a hardware level.

Abstract: Provided is a criteria generation device and the like that al low high-speed that achieves high-speed search even if a database contains unsearchable information. The criteria generation device reads, from a memory, target search criteria including first search criteria and second search criteria representing criteria for retrieving information from a database; calculates, regarding the read target search criteria, truth values indicating whether or not search criteria being a subset of the first search criteria are satisfied; sets the calculated truth values for the subset of the search criteria; and generates third search criteria depending on whether the second search criteria are true or false in accordance with whether or not the target search criteria for which the truth values are set are satisfied.

Abstract: A secure computation system configured to perform multi-party computation on a value of a predetermined function whose argument includes secret data, comprises a plurality of server apparatuses; wherein the plurality of server apparatuses, comprise: storage units that store shares that are bases over (of) a finite field generated by performing secret sharing on the secret data; share expansion units that generate extended shares by expanding the shares; OR operation units that perform OR operations included in the predetermined functions using the extended shares; and NOT operation units that perform NOT operations included in the predetermined functions using the extended shares.

Abstract: An intermediate apparatus that upon reception of a request from an application apparatus, instructs a plurality of secure computation apparatuses to perform a secret computation processing, in accordance with the request, performs a part of operation of the request from the application apparatus, on at least one of a part of data included in the request or data reconstructed from shares received from a plurality of secure computation apparatuses.

Abstract: A registration apparatus generates shares by secret sharing of a character string with a plurality of modulus and sends the shares to a plurality of server apparatuses to be stored therein. A retrieval apparatus sends shares generated by secret sharing of a retrieval character string with the plurality of modulus to the plurality of server apparatuses. The plurality of server apparatuses execute a subroutine for shares of the each registration character string stored in a storage unit and for each of the plurality of modulus, reconstruct an execution result, and determine whether or not to return the shares of the registration character string stored in the storage unit as a retrieval result. A retrieval apparatus reconstructs shares returned from the plurality of server apparatuses and obtains a retrieval result in which the retrieval character string hits, from the reconstructed result by the Chinese remainder theorem.

Abstract: A decoding apparatus includes a plurality-of-bits decoding part configured to receive an input vector obtained by adding a message encrypted by a trapdoor function and an error vector including an element(s) conforming with a discrete Gaussian distribution, and decode a plurality of bits from a lower bit of the message based on the input vector in correctness with a predetermined probability; and a confirmation calculation part configured to determine in parallel whether the decoded plurality of bits are correct or not, wherein the message is encrypted by taking an inner product with a vector including a power of two as an element(s).

Abstract: Provided is a database search device that, when searching an external database, efficiently executes any search command even if the usable search commands are restricted. The database search device comprises: a search command separating unit that separates input search commands into a first search command that can be executed in a prescribed database and a second search command that cannot be executed in the prescribed database; and a search command execution unit that provides the search results obtained by executing the second search command on the search results obtained by executing the first search command in the prescribed database, as the search results for the input search command.