This disclosure relates to a method for vertical federated learning. In multiple participation nodes deployed in a multi-way tree topology, an upper-layer participation node corresponds to k lower-layer participation nodes. After the upper-layer participation node and the k lower-layer participation nodes exchange public keys with each other, the upper-layer participation node performs secure two-party joint computation with the lower-layer participation nodes with a first public key and second public keys as encryption parameters to obtain k two-party joint outputs of a federated model. Further, the upper-layer participation node aggregates the k two-party joint outputs to obtain a first joint model output corresponding to the federated model. As such, a multi-way tree topology deployment-based vertical federated learning architecture is provided, improving the equality of each participation node in a vertical federated learning process.

The present disclosure generally relates to code verification. For example, aspects of the present disclosure include systems and techniques for determining whether two codes are a match. One example method generally includes generating, at a first device, first encrypted data at least in part by encrypting verification data using a public key; generating, at the first device, second encrypted data at least in part by encrypting a random factor using the public key; generating, at the first device, a key for the verification data; generating, at the first device, third encrypted data at least in part by encrypting the key using the public key; computing, at the first device, fourth encrypted data at least in part by applying homomorphic encryption function to the first encrypted data, the second encrypted data, and the third encrypted data; and sending, to a second device, the fourth encrypted data.

A homomorphic encryption-based testing computing system provides a risk-based, automated, one-directional push of production data through a homomorphic encryption tool and distributes the encrypted data to use in testing of applications. Data elements and test requirements are considered when automatically selecting a homomorphic encryption algorithm. A decisioning component selects an algorithm to use to homomorphically encrypt the data set and a push mechanism performs one or both of the homomorphic encryption and distribution of the encrypted data set to at least one intended host. Once delivered, the testing software and/or testing procedures proceed using the encrypted data set, where results of the testing may be stored in a data store. A validation mechanism may validate the test data against production data and communicates whether testing was successful.

Aspects of the present disclosure provide systems, methods, and computer-readable storage media that support secure training of machine learning (ML) models that preserves privacy in untrusted environments using distributed executable file packages. The executable file packages may include files, libraries, scripts, and the like that enable a cloud service provider configured to provide ML model training based on non-encrypted data to also support homomorphic encryption of data and ML model training with one or more clients, particularly for a diagnosis prediction model trained using medical data. Because the training is based on encrypted client data, private client data such as patient medical data may be used to train the diagnosis prediction model without exposing the client data to the cloud service provider or others. Using homomorphic encryption enables training of the diagnosis prediction model using encrypted data without requiring decryption prior to training.

Cryptographic-related processing is performed using an n-bit accelerator. The processing includes providing a binary operand to a multiply-and-accumulate unit of the n-bit accelerator. The multiply-and-accumulate unit performs an operation using the binary operand and a predetermined fractional constant F to obtain an operation result, and rounds the operation result by discarding x least-significant bits of the operation result to obtain a fractionally-scaled result, where x is a configurable number of bits to discard from the operation result, and the fractionally-scaled result facilitates performing the cryptographic-related processing.

According to an example, key splitting may include utilizing a masked version of a master key that is masked by using a mask.

A system and method generate private keys for devices participating in a self-certified identity based encryption scheme. A private key is used by the devices to establish a common session key for encoding digital communications between devices.

New and efficient protocols are provided for privacy-preserving machine learning training (e.g., for linear regression, logistic regression and neural network using the stochastic gradient descent method). A protocols can use the two-server model, where data owners distribute their private data among two non-colluding servers, which train various models on the joint data using secure two-party computation (2PC). New techniques support secure arithmetic operations on shared decimal numbers, and propose MPC-friendly alternatives to non-linear functions, such as sigmoid and softmax.

Some embodiments enable distributing data (e.g., recorded video, photographs, recorded audio, etc.) to a plurality of users in a manner which preserves the privacy of the respective users. Some embodiments leverage homomorphic encryption and proxy re-encryption techniques to manipulate the respective data so that selected portions of it are revealed according to an identity of the user currently accessing the respective data.

A system and method for secure cloud computing. The cloud based processing system comprises a user interface, allowing a user to enter and edit data, a proxy server, and a cloud based processing server. The user interface sends data entered by a user to the proxy server, which sends the encrypted data to the cloud based processing server. The proxy server receives editing commands from the user interface, and sends those commands to the cloud based processing server along with the encrypted data. The cloud based processing server receives the encrypted data and editing commands, applies the editing commands to the encrypted data, and sends the edited encrypted data back to the proxy server.