
Yuriy S. PolyakovDuality Technologies Inc.
Yuriy S. Polyakov
D.Sc. in Phys. & Math.
About
88
Publications
24,486
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
1,877
Citations
Introduction
I am a Principal Scientist at Duality Technologies, Inc. My current research interests are applied lattice-based cryptography, fully homomorphic encryption, and privacy-preserving machine learning. I am a co-founder and project lead of the OpenFHE homomorphic encryption library project. I am also a member of the Steering Committee for the HomomorphicEncryption.org standardization consortium. My research has been funded by DARPA, IARPA, NIH, and Sloan Foundation.
Publications
Publications (88)
Fully Homomorphic Encryption (FHE) is a cryptographic primitive that allows performing arbitrary operations on encrypted data. Since the conception of the idea in [RAD78], it has been considered a holy grail of cryptography. After the first construction in 2009 [Gen09], it has evolved to become a practical primitive with strong security guarantees....
Data privacy is a significant concern in many environments today. This is particularly true if sensitive information, e.g., engineering, medical, or financial data, is to be processed on potentially insecure systems, as it is often the case in cloud computing. Fully homomorphic encryption (FHE) offers a potential solution to this problem, as it all...
The notion that collaborative machine learning can ensure privacy by just withholding the raw data is widely acknowledged to be flawed. Over the past seven years, the literature has revealed several privacy attacks that enable adversaries to extract information about a model's training dataset by exploiting access to model parameters during or afte...
Real-world healthcare data sharing is instrumental in constructing broader-based and larger clinical datasets that may improve clinical decision-making research and outcomes. Stakeholders are frequently reluctant to share their data without guaranteed patient privacy, proper protection of their datasets, and control over the usage of their data. Fu...
Secure computation is of critical importance to not only the DoD, but across financial institutions, healthcare, and anywhere personally identifiable information (PII) is accessed. Traditional security techniques require data to be decrypted before performing any computation. When processed on untrusted systems the decrypted data is vulnerable to a...
Ring-Learning-with-Errors (RLWE) has emerged as the foundation of many important techniques for improving security and privacy, including homomorphic encryption and post-quantum cryptography. While promising, these techniques have received limited use due to their extreme overheads of running on general-purpose machines. In this paper, we present a...
A comparison of two encrypted numbers is an important operation needed in many machine learning applications, for example, decision tree or neural network inference/training. An efficient instantiation of this operation in the context of fully homomorphic encryption (FHE) can be challenging, especially when a relatively high precision is sought. Th...
The Cheon-Kim-Kim-Song (CKKS) homomorphic encryption scheme is currently the most efficient method to perform approximate homomorphic computations over real and complex numbers. Although the CKKS scheme can already be used to achieve practical performance for many advanced applications, e.g., in machine learning, its broader use in practice is hind...
The Brakerski-Gentry-Vaikuntanathan (BGV) and Brakerski/ Fan-Vercauteren (BFV) schemes are the two main homomorphic encryption (HE) schemes to perform exact computations over finite fields and integers. Although the schemes work with the same plaintext space, there are significant differences in their noise management, algorithms for the core homom...
We consider the problem of protecting image classifiers simultaneously from inspection attacks (i.e., attacks that have read access to all details in the program’s code) and black-box attacks (i.e., attacks where have input/output access to the program’s code). Our starting point is cryptographic program obfuscation, which guarantees some provable...
e18725
Background: Healthcare data sharing is important for the creation of diverse and large data sets, supporting clinical decision making, and accelerating efficient research to improve patient outcomes. This is especially vital in the case of real world data analysis. However, stakeholders are reluctant to share their data without ensuring pati...
Homomorphic encryption (HE) enables processing encrypted data without decrypting it. This technology can be used, for example, to allow a public cloud to operate on secret data without the cloud learning anything about the data. Simply encrypt the secret data with homomorphic encryption before sending it to the cloud, have the cloud process the enc...
Background:
Genome-Wide Association Studies (GWAS) refer to observational studies of a genome-wide set of genetic variants across many individuals to see if any genetic variants are associated with a certain trait. A typical GWAS analysis of a disease phenotype involves iterative logistic regression of a case/control phenotype on a single-neucloti...
Genome-wide association studies (GWASs) seek to identify genetic variants associated with a trait, and have been a powerful approach for understanding complex diseases. A critical challenge for GWASs has been the dependence on individual-level data that typically have strict privacy requirements, creating an urgent need for methods that preserve th...
Homomorphic Encryption (HE) is an emerging technology that enables computing on data while the data is encrypted. A major challenge with homomorphic encryption is that it takes extensive expert knowledge to design meaningful and useful programs that are constructed from atomic HE operations.
We present RAMPARTS to address this challenge. RAMPARTS p...
Many advanced lattice cryptography applications require efficient algorithms for inverting the so-called “gadget” matrices, which are used to formally describe a digit decomposition problem that produces an output with specific (statistical) properties. The common gadget inversion problems are the classical (often binary) digit decomposition, subga...
Homomorphic encryption is an emerging form of encryption that provides the ability to compute on encrypted data without ever decrypting them. Potential applications include aggregating sensitive encrypted data on a cloud environment and computing on the data in the cloud without compromising data privacy. There have been several recent advances res...
Advances in training models with log-linear structures, with topics including variable selection, the geometry of neural nets, and applications.
Log-linear models play a key role in modern big data and machine learning applications. From simple binary classification models through partition functions, conditional random fields, and neural nets, log...
In 2009, Craig Gentry introduced the first "fully" homomorphic encryption scheme allowing arbitrary circuits to be evaluated on encrypted data. Homomorphic encryption is a very powerful cryptographic primitive, though it has often been viewed by practitioners as too inefficient for practical applications. However, the performance of these encryptio...
Lattice trapdoors are an important primitive used in a wide range of cryptographic protocols, such as identity-based encryption (IBE), attribute-based encryption, functional encryption, and program obfuscation. In this paper, we present software implementations of the Gentry-Peikert-Vaikuntanathan (GPV) digital signature, IBE and ciphertext-policy...
We report on our implementation of a new Gaussian sampling algorithm for lattice trapdoors. Lattice trapdoors are used in a wide array of lattice-based cryptographic schemes including digital signatures, attributed-based encryption, program obfuscation and others. Our implementation provides Gaussian sampling for trapdoor lattices with prime moduli...
In this paper, we report on our implementation of a lattice-based Key-Policy Attribute-Based Encryption (KP-ABE) scheme, which uses short secret keys. The particular KP-ABE scheme can be used directly for Attribute-Based Access Control (ABAC) applications, as well as a building block in more involved applications and cryptographic schemes such as a...
We develop two IND-CPA-secure multihop unidirectional Proxy Re-Encryption (PRE) schemes by applying the Ring-LWE (RLWE) key switching approach from the homomorphic encryption literature. Unidirectional PRE is ideal for secure publish-subscribe operations where a publisher encrypts information using a public key without knowing upfront who the subsc...
Modern medical data information systems must collect and present data to authorized users. Distributed long-term medical data storage requires an effective security model for delegating data access. An effective delegation model must keep data encrypted at all times and avoid the need to share decryption keys. We present a secure information archit...
This article presents PICADOR, a system for end-to-end encrypted Publish-Subscribe information distribution with proxy re-encryption. The main novelty of PICADOR is that it provides an information distribution service with end-to-end encryption where publishers and subscribers do not need to establish shared encryption and decryption keys. Multiple...
The design of modern medical data information systems is driven by the need to collect and present data to authorized users. For collected medical data to be effective and improve patient treatment it must be transported from a device, aggregated, and analyzed to produce results that can be shared with care providers. Medical data may be analyzed a...
Embedded medical devices, such as wearable devices, are becoming increasingly common, but data from these devices is both very private and highly vulnerable to theft. Data needs to be collected from multiple devices to improve the effectiveness of treatment. The medical devices, data processing sites and intended care givers are often geographicall...
Elliptic curve cryptography (ECC) is an approach to public key cryptography (PKC) that is based on algebraic operations with elliptic curves defined over finite fields. Security of elliptic curve cryptography is based on the hardness of the elliptic curve discrete logarithm problem (ECDLP). Although there is no theoretical proof that ECDLP is intra...
The possibility of earthquake prediction is one of the key open questions in modern geophysics. We propose an approach based on the analysis of common short-term candidate precursors (2 weeks to 3 months prior to strong earthquake) with the subsequent processing of brain activity signals generated in specific types of rats (kept in laboratory setti...
A discrete multilayer model for the deposition of solutes inside membrane pores during ultrafiltration is proposed. The model takes into account the time-dependent steric exclusion of solutes at the pore inlet and the difference between the deposition coefficients for the first and higher layers caused by the action of the double-layer electrostati...
We use flicker-noise spectroscopy (FNS), a phenomenological method for the
analysis of time and spatial series operating on structure functions and power
spectrum estimates, to identify and study harmful chatter vibrations in a
regenerative turning process. The 3D cutting force components experimentally
measured during stainless steel turning are a...
Several recent studies have quantified the performance characteristics of ultrafiltration membranes in terms of the inherent trade-off between the membrane selectivity and permeability. However, none of these studies have accounted for the effects of membrane fouling on the evolution of the selectivity and permeability during typical ultrafiltratio...
Water scarcity in hot deserts, which cover about one-fifth of the Earth’s land area, along with rapid expansion of hot deserts into arable lands is one of the key global environmental problems. As hot deserts are extreme habitats characterized by the availability of solar energy with a nearly complete absence of organic life and water, space techno...
A mathematical model describing the performance of deadend outside-in hollow fiber filters developed on the basis of Darcy's law and the theory of depth filtration accounting for the decrease in the liquid flow rate with filter depth coordinate due to permeate withdrawal is studied. The system of governing equations is solved with the generalized C...
A phenomenological theory of depth membrane filtration (DMF), in which outside-in hollow fiber membranes are used as collectors of colloidal particles, is developed to study the performance of a hollow fiber membrane filter with two product streams, permeate (clarified liquid that passed through semipermeable membrane) and filtrate (clarified liqui...
Effect of transmembrane pressure (TMP), particle deposition and reentrainment coefficients, and specific cake resistance on the performance of a dead-end outside-in hollow fiber (DOIHF) filter and hollow fiber membrane (HFM) adsorber is studied using the adsorption–peptization model. It is shown that the increase in TMP causes a higher decline in t...
Two opposite design strategies for ultrafiltration/microfiltration filters: (1) reduction of concentration polarization and particle deposition to increase permeate velocity and (2) utilization of particle deposition on membrane surface to produce an additional (to permeate) volume of clarified water, are analyzed. It is shown that the first strate...
We propose a new type of earthquake precursor based on the analysis of correlation dynamics between geophysical signals of different nature. The precursor is found using a two-parameter cross-correlation function introduced within the framework of flicker-noise spectroscopy, a general statistical physics approach to the analysis of time series. We...
We examine stochastic variability in the dynamics of X-ray emission from the
black hole system GRS 1915+105, a strongly variable microquasar commonly used
for studying relativistic jets and the physics of black hole accretion. The
analysis of sample observations for 13 different states in both soft (low) and
hard (high) energy bands is performed by...
We apply flicker-noise spectroscopy (FNS), a time series analysis method
operating on structure functions and power spectrum estimates, to study the
clinical electroencephalogram (EEG) signals recorded in children/adolescents
(11 to 14 years of age) with diagnosed schizophrenia-spectrum symptoms at the
National Center for Psychiatric Health (NCPH)...
We use the anomalous-diffusion interpolation and parameterization procedure developed within the framework of flicker-noise spectroscopy (FNS) to characterize the stochastic variability of X-ray emission from accreting black hole GRS 1915+105. Our analysis of the lightcurves with a sampling frequency of 16 Hz in soft and hard X-ray energy bands sho...
A phenomenological systems approach for identifying potential precursors in
multiple signals of different types for the same local seismically active
region is proposed based on the assumption that a large earthquake may be
preceded by a system reconfiguration (preparation) at different time and space
scales. A nonstationarity factor introduced wit...
The functional properties of many technological surfaces in biotechnology, electronics, and mechanical engineering depend to a large degree on the individual features of their nanoscale surface texture, which in turn is a function of the surface manufacturing process. Among these features, the surface irregularities and self-similarity structures a...
The possibility in principle of extending metrological concepts to the characteristics of complex objects, the primary information
about the state or structure of which is presented in the form of complex chaotic dependences and cannot be expressed using
standard metrological images such as directly measured time and length and other dimensional va...
Closed bioregenerative life support systems: Applicability to hot deserts
A model of turbulent heat and mass transfer with nonlinear sources for the modeling of fast poly-merization processes under
matching conditions (equality of the values of temperatures and local flows) on the reactor—coolant interface was developed.
The finite-difference numerical solution was obtained via the method of alternating directions. The i...
We propose an interpolation expression using the difference moment (Kolmogorov transient structural function) of the second order as the average characteristic of displacements for identifying the anomalous diffusion in complex processes when the stochastic (the term "stochastic" in this paper refers to random variability in the signals of complex...
The effect of transmembrane pressure, pore depth and initial radius, inlet concentration of suspended particles, and particle-collection
efficiency of pore walls on the permeate rate and selectivity of ultraand microfiltration membranes with pores of the same
radius operated under standard blocking conditions is studied using the previously develop...
Water scarcity in hot deserts, which cover about one-fifth of the Earth’s land area, along with rapid expansion of hot deserts into arable lands is one of the key global environmental problems. This paper proposes and substantiates an interdisciplinary approach to sustainable settlements in hot deserts using the accomplishments in the design of clo...
In our earlier study dealing with the analysis of neuromagnetic responses (magnetoencephalograms - MEG) to flickering-color stimuli for a group of control human subjects (9 volunteers) and a patient with photosensitive epilepsy (a 12-year old girl), it was shown that Flicker-Noise Spectroscopy (FNS) was able to identify specific differences in the...
Evolution of Earth’s climate system over the past 800,000 years represents a complex process with successions of uneven glacial and interglacial periods. The length, amplitudes, and development of each climate cycle depend on a number of different factors, including the orbital parameters attributed to insolation and the complex responses of the Ea...
This review presents the fundamentals of Flicker-Noise Spectroscopy (FNS), a general phenomenological methodology in which the dynamics and structure of complex systems, characterized by nonlinear interactions, dissipation, and inertia, are analyzed by extracting information from various signals with stochastically varying components generated by t...
The flicker-noise spectroscopy (FNS) approach is used to determine the dynamic characteristics of neuromagnetic responses by analyzing the magnetoencephalographic (MEG) signals recorded as the response of a group of control human subjects and a patient with photosensitive epilepsy (PSE) to equiluminant flickering stimuli of different color combinat...
Hot deserts that presently cover about one-fifth of the land area of our planet are rapidly devouring more and more arable lands mostly due to anthropogenic causes. We propose an interdisciplinary approach to revitalizing and commercializing hot deserts, which is based on systems thinking and Russian and NASA space technology experience in designin...
A macroscopic depth filtration theory for the standard blocking of the pores of UF/MF membrane is developed. The theory predicts the membrane permeation rate and selectivity based on the introduction of the classical filter coefficient and shows that the profile of the number of deposited particles over the pore length is highly nonuniform (steep),...
The problem of information extraction from discrete stochastic time series, produced with some finite sampling frequency, using flicker-noise spectroscopy, a general framework for information extraction based on the analysis of the correlation links between signal irregularities and formulated for continuous signals, is discussed. It is shown that...
Macroscopic equations of the theory of filtration through granular beds are used to consider the spatial nonuniformity of
particle deposition on pore walls during the standard blocking of the pores of ultrafiltration (UF) and microfiltration (MF)
membranes. It is shown that the thickness of the layer of particles deposited inside the pore is charac...
A model of turbulent heat-and mass transfer with nonlinear sources (sinks) that appear as a result of chemical reactions of the first and second orders has been developed. The conjugation condition (equality of temperatures and local fluxes) at the reactor --- coolant interface was used for this purpose. A finite-difference numerical solution by th...
Fundamental principles and topics of a new approach to designing ultra- and microfiltration plants in which cake deposition
and its control are used to improve the membrane filtration efficiency are discussed. A general phenomenological mathematical
model for the process of depth membrane filtration, which was suggested by the new approach, is form...
Flicker-Noise Spectroscopy (FNS), a general approach to the extraction and parameterization of resonant and stochastic components contained in medical time series, is presented. The basic idea of FNS is to treat the correlation links present in sequences of different irregularities, such as spikes, "jumps", and discontinuities in derivatives of dif...
A mathematical model taking into account the adsorption of particles around a membrane pore mouth at the stage of standard
blocking is developed. The model is used to study the effect of the average membrane pore size, the cutoff coefficient, and
the ratio of the length of the pore inlet region to be blocked to the whole pore length on the product...
A generalized variable-parameter averaging (GVPA) method to solve nonlinear differential and integrodifferential equations is formulated. The method uses a constant-parameter solution to the original problem together with averaging and interpolation to obtain approximate solutions to some chemical engineering problems with variable parameters, such...
A transient model of heat and mass transfer with nonlinear sources (sinks) caused by first-and second-order chemical reactions is developed. The model uses a matching condition (equal temperature and local flux values) at the reaction zone-coolant interface. A finite-difference numerical solution to the problem is obtained using the alternating dir...
A mathematical model of cake formation in deadend outside-in hollow fiber filters accounting for nonuniform cake deposition
throughout the filter is developed. The model is used to study two possible mechanisms of cake growth: conventional (convective)
and adsorption-peptization. The system of equations describing the cake formation is numerically...
The effect of transmembrane pressure, feed flow rate, and the geometry of hollow-fiber filters on their performance is studied for depth membrane filtration with reversible adsorption. Mathematical models for rectangular and radial filters are used to find the maximal operation times with a filter retention of not less than 0.9 at constant product...
A novel membrane separation process in which outside-in hollow fiber (HF) membranes with axes normal to the flow of the suspension under treatment are used as collectors of the colloidal particles is proposed. This process, called depth membrane filtration, yields two product streams: permeate, the liquid that passed through the semipermeable HF me...
A mathematical model of a depth filter that removes colloidal particles from suspension by reversible adsorption is studied. The filtration problem is analytically solved when the adsorption and peptization coefficients are written as a function of van der Waals, electrical double layer, and Born interaction potential energies. The solution is used...
A new accelerated algorithm to solve the single-facility minisum location problem is developed. The acceleration is achieved using a feedback factor. The proposed algorithm converges faster than the accelerating procedures available in the literature. Being nearly as simple as the classical Weiszfeld procedure, the new method can easily be implemen...
Microfiltration is described by a mathematical model taking into account complete pore blocking, gradual pore blocking with an increase in the retentivity of the microfilter (standard blocking), primary layer (sublayer) formation, and cake formation. Membrane porosity is described by a lognormal pore-size distribution. The convexity of the kinetic...
General idea of nonlinear correcting factors that was successfully applied to accelerate and derive algorithms for many linear and nonlinear problems is used to develop several efficient procedures for solving linear systems of equations based on