Microsoft Technology Licensing, LLC

Nishanth Chandran

Ramarathnam R Venkatesan

Satyanarayana V Lokam

Divya Gupta

Ramachandran Ramjee

Panagiotis Antonopoulos

Ranjita Bhagwan

Srinath T V Setty

Jonathan Alexander Lee

Kristin E Lauter

Michael James Zwilling

Rahul Anand Sharma

Saikrishna Badrinarayanan

Muthian Sivathanu

Srinath Setty

Melissa Erin Chase

Ulrich Homann

Vinod Vaikuntanathan

Srinath Tumkur Venkatacha Setty

Apurv Mehra

Sameer Wagh

Daniel John Carroll, Jr

Christoph Berlin

Antonios Papadimitriou

Aseem Rastogi

Brant Lee Zwiefel

Sharmila Deva Selvis

Sambhav Satija

Jason Sydney Barnwell

Harmeet Singh

Mayank Rathee

Nishant Kumar

Sudheesh Singanamalla

Deevashwer Rathee

Basil Cherian

Embodiments described herein are directed to a verifiable identity map that maintains identities and public keys associated with the identities. The map is maintained by a ledger database that provides tamper-resistant/evident capabilities for tables (comprising the map) thereof. For instance, when a materialized view of the database is generated, the database provides a digest representative of a state thereof to computing devices that access the map for the keys. When the database receives a request from a device to access the map, the digest is received along therewith. The database is validated based on the digest to determine whether the database has been tampered with since the provision of the digest. Responsive to a successful validation, the database provides access in accordance with the request. When a key in the map is updated, the database subsequently generates a new digest, which is provided to the computing device.

Verifiable identity map maintaining identities and associated public keys

An entity is enabled to access encrypted resources in response to verifying access criteria of a region-based security policy is met. For example, a resource request to access an encrypted resource is received from an entity. A determination that the encrypted resource is assigned to a first region and is protected by a region-based security policy is made. A proof of a region attribute indicating that the entity possesses the region attribute is received from the entity, the region attribute indicates the entity is associated with the first region. An encrypted version of the region attribute is obtained from a ledger database. The resource request is validated based at least on the encrypted attribute and the proof of the region attribute. A verification is made that an access criteria of the region-based security policy is met. The entity is provided access to the encrypted resource.

Region-based security policies for cloud resources

Embodiments described herein enable at least one of a plurality of entities to access data protected by a security policy in response to validating respective digital access requests from the entities. The respective digital access requests are received, each comprising a proof. For each request, an encrypted secret share is obtained from a respective ledger database. Each request is validated based at least on the respective encrypted secret share and the proof, without decrypting the respective encrypted secret share. In response to validating all of the requests, a verification that an access criteria of a security policy is met is made. If so, at least one of the entities is provided with access to data protected by the security policy. In an aspect, embodiments enable a blind subpoena to be performed. In another aspect, embodiments enable the at least one entity to access the data for an isolated purpose.

Blind subpoena protection

A decryption key is recovered that is utilized to decrypt an encrypted resource. One or more location attribute policy (LAP) servers determine whether a user attempting to access a resource has the necessary attributes to access the resource and is in a valid location in which the user is required to be to access the resource. The attributes and location are defined by a policy assigned to the resource. To verify that the user has the required attributes, the LAP server(s) request a cryptographic proof from the user that proves that the user has the required attributes. Upon validating the proof, a first portion of the decryption key is released. The LAP server(s) release a second portion of the decryption key after verifying that the user is in the required location. The LAP server(s) generate the decryption key based on the released portions.

Decryption key generation and recovery

A secure inference over Deep Neural Networks (DNNs) using secure two-party computation to perform privacy-preserving machine learning. The secure inference uses a particular type of comparison that can be used as a building block for various layers in the DNN including, for example, ReLU activations and divisions. The comparison securely computes a Boolean share of a bit representing whether input value x is less than input value y, where x is held by a user of the DNN, and where y is held by a provider of the DNN. Each party computing system parses their input into leaf strings of multiple bits. This is much more efficient than if the leaf strings were individual bits. Accordingly, the secure inference described herein is more readily adapted for using in complex DNNs.

Private inference in deep neural network

A system includes a set of low resource devices, each configured to receive transactions to be added to an encrypted block chain ledger from a sample of untrusted high resource devices, prepare a proposed block of the received transactions, provide the proposed block to the sample of untrusted high resource devices, receive proposed blocks from the untrusted high resource devices originating from the set of low resource devices. The low resource devices run a consensus protocol to select one proposed block to add to the encrypted block chain ledger stored on the untrusted high resource devices.

Lightweight blockchain based on split-trust

In some embodiments, an encryption system secures data using a homomorphic encryption. The encryption system encrypts a number by encrypting a number identifier of the number and combining the number and the encrypted number identifier using a mathematical operation to generate an encrypted number. The encrypted numbers may be stored at a server system along with their number identifiers. The server system can then generate an aggregation (e.g., sum) of the encrypted numbers and provide the aggregation, the encrypted numbers, and the number identifiers. The encryption system can then separate the aggregation of the numbers from the aggregation of the encrypted numbers using an inverse of the mathematical operation used in the encryption to effect removal of an aggregation of the encrypted number identifiers of the numbers from the aggregation of the encrypted numbers. The separated aggregation of the numbers is an aggregation of the plurality of the numbers.

Aggregation of encrypted data

Systems and methods for private deep neural network training are disclosed. Method includes storing first private values at first machine and second private values at second machine; providing, to third machine, first share of first private values and first share of second private values; providing, to fourth machine, second share of first private values and second share of second private values; computing, at third machine, third machine-value based on first share of first private values and first share of second private values; computing, at fourth machine, fourth machine-value based on second share of first private values and second share of second private values; providing, to first machine and second machine, third machine-value and fourth machine-value; and computing, at first machine, a mathematical function of first private values and second private values, mathematical function being computed based on first private values stored at first machine, third machine-value, and fourth machine-value.

Private deep neural network training

A device establishes a key recovery policy and generates a key that is protected based on the key recovery policy. The key recovery policy indicates which combinations of other entities can recover the protected key. The device generates different shares of the protected key, each share being a value that, in combination with the other share(s), allows the protected key to be recovered. Each share is associated with a particular leaf agent, the device encrypts each share with the public key of the leaf agent associated with the share and provides the encrypted share to a service. When recovery of the protected key is desired, a recovering authority can generate the protected key only if the recovering authority receives decrypted shares from a sufficient one or combination of leaf agents as indicated by the recovery policy.

Policy-based key recovery

An encryption system stores encrypted values for aggregation is provided. The encryption system accesses an input set with input values. For each distinct value in the input set of input values, the encryption system generates an output set with an encrypted output value corresponding to each input value. The encryption system sets the encrypted output value for a corresponding input value to an encryption of an indicator of a match when the corresponding input value is the same as that distinct value. Otherwise, the encryption sets the encrypted output value for the corresponding input value to an encryption of an indicator of no match. The encrypted output values can then be aggregated to generate an encrypted aggregation based on input values that match, and the encrypted aggregation can be decrypted to generated a decrypted aggregation based on the input values that match.

Aggregation based on splayed data

An obfuscated policy data encryption system and method for re-encrypting data to maintain the confidentiality and integrity of data about a user when the data is stored in a public cloud computing environment. The system and method allow a user to specify in a data-sharing policy who can obtain the data and how much of the data is available to them. This policy is obfuscated such that it is unintelligible to the cloud operator and others processing and storing the data. In some embodiments, a patient species with whom his health care data should be shared with and the encrypted health care data is stored in the cloud in an electronic medical records system. The obfuscated policy allows the electronic medial records system to dispense the health care data of the patient to those requesting the data without disclosing the details of the policy itself.

User-controlled data encryption with obfuscated policy

Growing community of inventors

Bengaluru, India

Nishanth Chandran