Techniques are disclosed relating to the execution of machine learning models on client devices, particularly in the context of transaction risk evaluation. This reduces computational burden on server systems. In various embodiments, a server system may receive, from a client device, a request to perform a first operation and select a first machine learning model, from a set of machine learning models, to send to the client device. In some embodiments the first machine learning model is executable, by the client device, to generate model output data for the first operation based on one or more encrypted input data values that are encrypted with a cryptographic key inaccessible to the client device. The server system may send the first machine learning model to the client device and then receive, from the client device, a response message that indicates whether the first operation is authorized based on the model output data.

Disclosed are techniques for determining data relationships between privacy-restricted datapoints, sourced over a computer network, which require data privacy measures concealing at least some datapoints from other clients in the network that the datapoint respectively do not originate from. A first client encrypts a first datapoint with a public key of a public/private encryption scheme and communicates it to the second client along with the public key. The second client encrypts a corresponding second datapoint with the public key, then determines a relationship between the two encrypted datapoints, and communicates the determined relationship to a central client along with the public key. Random noise is encrypted by the central client and added to the determined relationship, then sent together to the first client, followed by decryption by the first client using the private key. The central client extracts the random noise after receiving the decrypted determined relationship.

Methods and systems for performing an operation on at least one homomorphically encrypted ciphertext, the method include determining, by a computing device, a value that is an initial approximation of a result of the operation on the at least one homomorphically encrypted ciphertext; and iteratively improving, by the computing device, the value using a recurrence relation wherein a number of iterations is determined based on a predetermined accuracy to minimize an approximation error.

Methods and system for risk determination and risk categorization using encrypted data are provided. The risk determination can involve determining an inner product operation between a generalized weight table and an encrypted incidence vector, summing the result of the inner product operation and/or decrypting the results. Method and systems for encrypting data for use in homomorphic risk determination are also provided.

Method for online advertising auction on a peer-to-peer network includes: deploying a smart contract to publish a need from a consumer; receiving encrypted ad bids by the smart contract; storing the received ad bids in a hash function; reducing a number of ads that can be displayed by the consumer; transmitting the ad price to the consumer via the peer-to-peer computer network, when the hashed verification code is received from the consumer verifying that the consumer has viewed the ad content within an ad-viewing period of time; transmitting a difference between the advance payment and the ad price to the advertiser, by the smart contract via the peer-to-peer network; and ending the online advertising auction.

Systems and techniques for data creation limits are described herein. In an example, a data creation limits system is adapted to receive data and split the data into a plurality of portions based on entity interests in each of the plurality of portions. The data creation limits system may be further adapted to generate respective tokens for each portion of the plurality of portions. The data creation limits system may be further adapted to assign an owner to a token of the respective tokens, the token corresponding to a portion of the plurality of portions and assigning the owner based on the owner having an entity interest in creation of the portion. The data creation limits system may be further adapted to generate a script, using the token, for access to the portion. The data creation limits system may be further adapted to save the portion including the token.

A system and method are disclosed for providing a quantum proof key exchange. The method includes generating at a first computing device a random bit a.sub.i, encrypting a.sub.i using quantum-proof homomorphic encryption ξ to yield ξ.sub.A(a.sub.i), transmitting ξ.sub.A(a.sub.i) to a second computing device, generating at the second computing device a random bit b.sub.i, encrypting b.sub.i using the quantum-proof homomorphic encryption ξ to yield ξ.sub.B(b.sub.i), transmitting ξ.sub.B(b.sub.i) to the first computing device and generating a common key between the first computing device and the second computing device based on ξ.sub.A(a.sub.i) and ξ.sub.B(b.sub.i).

A method comprises receiving an ad event data including data about a plurality of ad events, and including a user ID and an ad ID for each ad event in the ad event data set, where the ad event data set has been anonymized applying a one-way encryption key for each user ID in the ad event data set, and a two-way encryption key for the ad ID in the ad event data set. The attribution processor receives a customer data set including data about a plurality of customers, including a user ID and a customer value for each customer, where the customer data set has been anonymized using the one-way encryption key for each user ID in the data, and a private encryption key for the customer value. Without decrypting the received ad event data set and the received customer data set, the processor then matches ad events for each conversion by comparing the user IDs in the encrypted ad event data set to the user IDs in the encrypted customer data set to create a set of contributing ad events, assigns a share of the customer value to each relevant ad event, sums homomorphically the encrypted customer values for contributing events, and determines a recommendation for serving advertisements.

Embodiments of the present disclosure provide hierarchical, differential privacy enhancements to federated, machine learning. Local machine learning models may be generated and/or trained by data owners participating in the federated learning framework based on their respective data sets. Noise corresponding to and satisfying a first privacy loss requirement are introduced to the data owners' respective data sets, and noise corresponding to and satisfying a first privacy loss requirement are introduced to the local models generated and/or trained by the data owners. The data owners transmit model data corresponding to their respective local models to a coordinator, which in turn aggregates the data owners' model data. After introducing noise corresponding to and satisfying a third privacy loss requirement to the aggregated model data, the coordinator transmits the aggregated model data to the data owners to facilitate updating and/or re-training on their respective machine learning models.

Provided are an electronic device for sorting homomorphic ciphertext by using shell sorting and an operating method thereof to sort ciphertext generated by using homomorphic encryption according to a size of an original number corresponding thereto.