A system and method for managing medical information with enhanced personal information protection are disclosed. The system and method may generate a distributed key including a user key, an agent key, and a backup key, and manage medical data of a user, using a user agent assigned to each user and the distributed key, to prevent personal information of the user from being directly accessed from outside.

Methods for secure random selection of t client devices from a set of N client devices and methods for secure computation of inputs of t client devices randomly selected from N client devices are described. Such random selection method may include determining an initial binary vector b of weight t by setting the first t bits to one: b.sub.i=1, 1≤i≤t, and all further bits to zero: b.sub.i=0, t<i≤N; each client device i (i=1, . . . , N) of the set of N client devices jointly generating a random binary vector b of weight t in an obfuscated domain on the basis of the initial binary vector b including: determining a position n in the binary vector; determining a random number r in {n, n+1, . . . N}; and, using the random number to swap binary values at positions n and r of the binary vector b.

Respective sets of homomorphically encrypted training data are received from multiple users, each encrypted by a key of a respective user. The respective sets are provided to a combined machine learning model to determine corresponding locally learned outputs, each in an FHE domain of one of the users. Conversion is coordinated of the locally learned outputs in the FHE domains into an MFHE domain, where each converted locally learned output is encrypted by all of the users. The converted locally learned outputs are aggregated into a converted composite output in the MFHE domain. A conversion is coordinated of the converted composite output in the MFHE domain into the FHE domains of the corresponding users, where each converted decrypted composite output is encrypted by only a respective one of the users. The combined machine learning model is updated based on the converted composite outputs. The model may be used for inferencing.

A method, apparatus and system for providing controlled access to data in a distributed computing environment include storing received data to be accessed via the distributed computing environment in at least one storage device, generating at least one integrity data structure identifying at least a storage location of at least a respective portion of the stored data, storing the generated at least one integrity data structure in a block of a blockchain, encrypting the at least one integrity data structure in the block of the blockchain, and selectively providing at least a portion of at least one decryption key for decrypting the encrypted at least one integrity data structure to enable access to the respective portion of the stored data for which the at least one integrity data structure is generated. Additionally, the stored data can be encrypted and a decryption key can be provided for decrypting the stored data.

A system and method is provided for providing distributed computing platform on untrusted hardware. An exemplary method includes launching a hypervisor on an untrusted computing node and receiving a request generated to provide a computing function using hardware of the untrusted computing node. Upon receiving the request, an enclave in memory of the untrusted computing node is created and a virtual machine is launched in the memory enclave. Moreover, a guest operating system of the virtual machine verifies the security of the untrusted computing node. Finally, the guest operating system performs the computing function using the hardware of the untrusted computing node upon the guest operating system verifying the security of the untrusted computing node and the hypervisor.

A method of performing ordered statistics between at least two parties is disclosed which includes identifying a first dataset (x.sub.A) by a first node (A), identifying a second dataset (x.sub.B) by a second node (B), wherein x.sub.B is unknown to A and x.sub.A is unknown to B, and wherein A is in communication with B, and wherein A and B are in communication with a server (S), A and B each additively splitting each member of their respective datasets into corresponding shares, sharing the corresponding shares with one another, arranging the corresponding shares according to a mutually agreed predetermined order into corresponding ordered shares, shuffling the ordered shares into shuffled shares, re-splitting the shuffled shares into re-split shuffled shares, and performing an ordered statistical operation on the re-split shuffled shares, wherein the steps of shuffle and re-split is based on additions, subtractions but not multiplication and division.

A computerized system and method for symmetric encryption and decryption using two machines, the method including obtaining a message and an initialization vector on a first machine, sending the initialization vector to a second machine, where said second machine stores an encryption key for a Key Derivation Function (KDF), generating a derived key on the second machine by applying the KDF receiving as input both the encryption key and the initialization vector, sending the derived key from the second machine to the first machine, and encrypting the message using the derived key on the first machine.

A method of providing blind vertical learning includes creating, based on assembled data, a neural network having n bottom portions and a top portion and transmitting each bottom portion of then bottom portions to a client device. The training of the neural network includes accepting a, output from each bottom portion of the neural network, joining the plurality of outputs at a fusion layer, passing the fused outputs to the top portion of the neural network, carrying out a forward propagation step at the top portion of neural network, calculating a loss value after the forward propagation step, calculating a set of gradients of the loss value with respect to server-side model parameters and passing subsets of the set of gradients to a client device. After training, the method includes combining the trained bottom portion from each client device into a combined model.

Multiple systems, methods, and computer program product embodiments for password-less authentication using key agreement and multi-party computation (MPC). In one or more embodiments, following an authentication request received by a host computing device, the host computing device and a user computing device generate a shared key using a key agreement algorithm. Then, the host computing device generates a challenge that is encrypted using the shared key and transmitted to the user computing device. The user computing device decrypts the challenge after regenerating the shared key and sends the decrypted result to the host computing device as the challenge response. The authentication request is granted by the host computing device if the challenge and the challenge response match. New keys and a new challenge are generated for each authentication request. This process relies on public key cryptography eliminating the needs for passwords.

A homomorphic inference device (500) divides a model ciphertext into a ciphertext for inference and a ciphertext for computation, generates a preliminary result ciphertext by a homomorphic operation algorithm without decrypting the ciphertext for computation and a data ciphertext, and generates an inference result ciphertext by a homomorphic operation algorithm, using the preliminary result ciphertext and the ciphertext for inference that have not been decrypted. A partial decryption device (600) generates a partial decryption result by performing partial decryption on the inference result ciphertext, using a model secret key. A final decryption device (700) decrypts an inference result from the partial decryption result, using a data secret key.