The accelerating convergence of Cloud computing and the Internet of Things (IoT) has revolutionized data-driven services, yet it has also introduced a significant trust deficit in highly regulated sectors such as healthcare and finance. Traditional architectures, characterized by static security protocols and reactive monitoring, are increasingly inadequate for protecting sensitive medical records and financial assets against sophisticated cyber-threats and operational anomalies. This review article proposes a "Cognitive Cloud–IoT Architecture" that integrates human-like reasoning, self-learning, and context-aware decision-making into the data exchange process. We evaluate a multi-layered framework comprising an intelligent perception layer, a cognitive middleware reasoning engine, and a secure cloud core designed to establish objective trust through continuous verification. The study analyzes key mechanisms for trustworthy exchange, including Zero-Knowledge Proofs (ZKP), blockchain-enabled immutable ledgers, and privacy-preserving federated learning. In the healthcare domain, we examine the application of "cognitive patients" through remote monitoring systems that differentiate between sensor noise and clinical emergencies. In the financial sector, we explore the "cognitive ledger" for autonomous fraud forensics and secure cross-border settlements. Furthermore, the article addresses critical strategic challenges, such as the computational overhead of running cognitive models on edge devices and the legal necessity of algorithmic explainability. By synthesizing future trends, including quantum-safe hybridization and sovereign cognitive clouds, this research provides a comprehensive roadmap for developing resilient, intelligent ecosystems. Ultimately, we demonstrate that cognitive architecture is the essential bridge to an "invisible intelligence" that ensures the integrity of human life and global financial stability in an increasingly connected world.
Cai, Dongliang, Borui Chen, Liang Zhang, Haibin Kan
Attribute-based encryption (ABE) is a generalization of public-key encryption that enables fine-grained access control in cloud services. Recently, Hohenberger et al. (Eurocrypt 2023) introduced the notion of registered ABE, which is an ABE scheme without a trusted central authority. Instead, users generate their own public/secret keys and then register their keys and attributes with a key curator. The key curator is a transparent and untrusted entity and its behavior needs to be audited for malicious registration. In addition, pairing-based registered ABE still suffers the heavy decryption overhead like ABE. A general approach to address this issue is to outsource decryption to a decryption cloud service (DCS).In this work, we propose BA-ORABE, the first fully auditable registered ABE with reliable outsourced decryption scheme based on blockchain. First, we utilize a verifiable tag mechanism to achieve verifiability of ciphertext transformation, and the exemptibility which enables the honest DCS to escape from wrong claims is guaranteed by zero knowledge fraud proof under optimistic assumption. Additionally, our system achieves fairness and decentralized outsourcing to protect the interests of all parties and the registration and outsourcing process are transparent and fully auditable through blockchain. Finally, we give security analysis, implement and evaluate our scheme on Ethereum to demonstrate its feasibility and efficiency, and show its advantages in real application of decentralized finance.
5G networks provide secure and reliable information transmission services for the Internet of Everything, thus paving the way for 6G networks, which is anticipated to be an AI-based network, supporting unprecedented intelligence across applications. Abundant computing resources will establish the 6G Computing Power Network (CPN) to facilitate ubiquitous intelligent services. In this article, we propose BECS, a computing sharing mechanism based on evolutionary algorithm and blockchain, designed to balance task offloading among user devices, edge devices, and cloud resources within 6G CPN, thereby enhancing the computing resource utilization. We model computing sharing as a multi-objective optimization problem, aiming to improve resource utilization while balancing other issues. To tackle this NP-hard problem, we devise a kernel distance-based dominance relation and incorporated it into the Non-dominated Sorting Genetic Algorithm III, significantly enhancing the diversity of the evolutionary population. In addition, we propose a pseudonym scheme based on zero-knowledge proof to protect the privacy of users participating in computing sharing. Finally, the security analysis and simulation results demonstrate that BECS can fully and effectively utilize all computing resources in 6G CPN, significantly improving the computing resource utilization while protecting user privacy.
Zero-knowledge proof (ZKP) is an attractive cryptographic paradigm that allows a party to prove the correctness of a given statement without revealing any additional information. It offers both computation integrity and privacy, witnessing many celebrated deployments, such as computation outsourcing and cryptocurrencies. Recent general-purpose ZKP schemes, e.g., zero-knowledge succinct non-interactive argument of knowledge (zk-SNARK), suffer from time-consuming proof generation, which is mainly bottlenecked by the large-scale number theoretic transformation (NTT) and multi-scalar point multiplication (MSM). To boost its wide application, great interest has been shown in expediting the proof generation on various platforms like GPU, FPGA and ASIC.So far as we know, current works on the hardware designs for ZKP employ two separated data-paths for NTT and MSM, overlooking the potential of resource reusage. In this work, we particularly explore the feasibility and profit of implementing both NTT and MSM with a unified and high-performance hardware architecture. For the crucial operator design, we propose a dual-precision, load-balanced and fully-pipelined Montgomery multiplier (LBFP MM) by introducing the new mixed-radix technique and improving the prior quotient-decoupled strategy. Collectively, we also integrate orthogonal ideas to further enhance the performance of LBFP MM, including the customized constant multiplication, truncated LSB/MSB multiplication/addition and Karatsuba technique. On top of that, we present the unified, scalable and highperformance hardware architecture that conducts both NTT and MSM in a versatile pipelined execution mechanism, intensively sharing the common computation and memory resource. The proposed accelerator manages to overlap the on-chip memory computation with off-chip memory access, considerably reducing the overall cycle counts for NTT and MSM.We showcase the implementation of modular multiplier and overall architecture on the BLS12-381 elliptic curve for zk-SNARK. Extensive experiments are carried out under TSMC 28nm synthesis and similar simulation set, which demonstrate impressive improvements: (1) the proposed LBFP MM obtains 1.8x speed-up and 1.3x less area cost versus the state-of-the-art design; (2) the unified accelerator achieves 12.1x and 5.8x acceleration for NTT and MSM while also consumes 4.3x lower overall on-chip area overhead, when compared to the most related and advanced work PipeZK.
Abstract We propose a generic compiler that can convert any zero-knowledge (ZK) proof for SIMD circuits to general circuits efficiently, and an extension that can preserve the space complexity of the proof systems. Our compiler can immediately produce new results improving upon state of the art. By plugging in our compiler to Antman, an interactive sublinear-communication protocol, we improve the overall communication complexity for general circuits from $$\mathcal {O}(C^{3/4})$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>O</mml:mi> <mml:mo>(</mml:mo> <mml:msup> <mml:mi>C</mml:mi> <mml:mrow> <mml:mn>3</mml:mn> <mml:mo>/</mml:mo> <mml:mn>4</mml:mn> </mml:mrow> </mml:msup> <mml:mo>)</mml:mo> </mml:mrow> </mml:math> to $$\mathcal {O}(C^{1/2})$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>O</mml:mi> <mml:mo>(</mml:mo> <mml:msup> <mml:mi>C</mml:mi> <mml:mrow> <mml:mn>1</mml:mn> <mml:mo>/</mml:mo> <mml:mn>2</mml:mn> </mml:mrow> </mml:msup> <mml:mo>)</mml:mo> </mml:mrow> </mml:math> . Our implementation shows that for a circuit of size $$2^{27}$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:msup> <mml:mn>2</mml:mn> <mml:mn>27</mml:mn> </mml:msup> </mml:math> , it achieves up to $$83.6\times $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>83.6</mml:mn> <mml:mo>×</mml:mo> </mml:mrow> </mml:math> improvement on communication compared to the state-of-the-art implementation. Its end-to-end running time is at least $$70\%$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mn>70</mml:mn> <mml:mo>%</mml:mo> </mml:mrow> </mml:math> faster in a 10Mbps network. Using the recent results on compressed $$\varSigma $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>Σ</mml:mi> </mml:math> -protocol theory, we obtain a discrete-log-based constant-round zero-knowledge argument with $$\mathcal {O}(C^{1/2})$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>O</mml:mi> <mml:mo>(</mml:mo> <mml:msup> <mml:mi>C</mml:mi> <mml:mrow> <mml:mn>1</mml:mn> <mml:mo>/</mml:mo> <mml:mn>2</mml:mn> </mml:mrow> </mml:msup> <mml:mo>)</mml:mo> </mml:mrow> </mml:math> communication and common random string length, improving over the state of the art that has linear-size common random string and requires heavier computation. We improve the communication of a designated n -verifier zero-knowledge proof from $$\mathcal {O}(nC/B+n^2B^2)$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>O</mml:mi> <mml:mo>(</mml:mo> <mml:mi>n</mml:mi> <mml:mi>C</mml:mi> <mml:mo>/</mml:mo> <mml:mi>B</mml:mi> <mml:mo>+</mml:mo> <mml:msup> <mml:mi>n</mml:mi> <mml:mn>2</mml:mn> </mml:msup> <mml:msup> <mml:mi>B</mml:mi> <mml:mn>2</mml:mn> </mml:msup> <mml:mo>)</mml:mo> </mml:mrow> </mml:math> to $$\mathcal {O}(nC/B+n^2)$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>O</mml:mi> <mml:mo>(</mml:mo> <mml:mi>n</mml:mi> <mml:mi>C</mml:mi> <mml:mo>/</mml:mo> <mml:mi>B</mml:mi> <mml:mo>+</mml:mo> <mml:msup> <mml:mi>n</mml:mi> <mml:mn>2</mml:mn> </mml:msup> <mml:mo>)</mml:mo> </mml:mrow> </mml:math> . To demonstrate the scalability of our compilers, we were able to extract a commit-and-prove SIMD ZK from Ligero and cast it in our framework. We also give one instantiation derived from LegoSNARK, demonstrating that the idea of CP-SNARK also fits in our methodology.
Data sharing is ubiquitous in the metaverse, which adopts blockchain as its foundation. Blockchain is employed because it enables data transparency, achieves tamper resistance, and supports smart contracts. However, securely sharing data based on blockchain necessitates further consideration. Ciphertext-policy attribute-based encryption (CP-ABE) is a promising primitive to provide confidentiality and fine-grained access control. Nonetheless, authority accountability and key abuse are critical issues that practical applications must address. Few studies have considered CP-ABE key confidentiality and authority accountability simultaneously. To our knowledge, we are the first to fill this gap by integrating non-interactive zero-knowledge (NIZK) proofs into CP-ABE keys and outsourcing the verification process to a smart contract. To meet the decentralization requirement, we incorporate a decentralized CP-ABE scheme into the proposed data sharing system. Additionally, we provide an implementation based on smart contract to determine whether an access control policy is satisfied by a set of CP-ABE keys. We also introduce an open incentive mechanism to encourage honest participation in data sharing. Hence, the key abuse issue is resolved through the NIZK proof and the incentive mechanism. We provide a theoretical analysis and conduct comprehensive experiments to demonstrate the feasibility and efficiency of the data sharing system. Based on the proposed accountable approach, we further illustrate an application in GameFi, where players can play to earn or contribute to an accountable DAO, fostering a thriving metaverse ecosystem.
Yanming Zhu, Xuefei Yin, Alan Wee‐Chung Liew, Hui Tian
With the rapid advancement of artificial intelligence and deep learning, medical image analysis has become a critical tool in modern healthcare, significantly improving diagnostic accuracy and efficiency. However, AI-based methods also raise serious privacy concerns, as medical images often contain highly sensitive patient information. This review offers a comprehensive overview of privacy-preserving techniques in medical image analysis, including encryption, differential privacy, homomorphic encryption, federated learning, and generative adversarial networks. We explore the application of these techniques across various medical image analysis tasks, such as diagnosis, pathology, and telemedicine. Notably, we organizes the review based on specific challenges and their corresponding solutions in different medical image analysis applications, so that technical applications are directly aligned with practical issues, addressing gaps in the current research landscape. Additionally, we discuss emerging trends, such as zero-knowledge proofs and secure multi-party computation, offering insights for future research. This review serves as a valuable resource for researchers and practitioners and can help advance privacy-preserving in medical image analysis.
What will computer security look like in the year 2100? This talk will begin with a few predictions that aim to suggest a few research directions in the present. We will then transition to the exciting area of applied zero knowledge proofs, an area that has seen tremendous growth in recent years. We will describe some of the new ideas in the space and focus on a number of remarkable real-word applications of these techniques. The talk will be self contained and accessible to all.
Alex Berke, Tobin South, Robert Mahari, Kent Larson · 5 authors
Tax returns contain financial information of interest to third parties: public officials are asked to share financial data for transparency, companies seek to assess the financial status of business partners, and individuals need to prove their income to third-parties.Tax returns also contain sensitive data such that sharing them in their entirety undermines privacy.We outline how zero-knowledge cryptography may be applied to address this tension by allowing individuals and organizations to make provable claims about select information in their tax returns without revealing additional information, in a way that can be independently verified by third parties.We highlight key system goals and design specifications for this zero-knowledge tax disclosure system (zkTax) and present a prototype implementation.The prototype consists of three distinct services that can be distributed: a tax authority that provides signed tax documents; a Redact & Prove Service that enables users to redact tax documents and produce a zero-knowledge proof attesting the provenance of the redacted data; and a Verify Service to check the validity of claims.We demonstrate how zkTax could be implemented with minimal changes to existing tax infrastructure, allowing the system to be extensible to other contexts and jurisdictions.This work provides a practical example of how distributed tools leveraging cryptography can enhance existing government or financial infrastructures, providing immediate transparency alongside privacy without system overhauls.
Yibin Yang, David Heath, Carmit Hazay, Vladimir Kolesnikov · 5 authors
We explore Zero-Knowledge Proofs (ZKPs) of statements expressed as programs written in high-level languages, e.g., C or assembly. At the core of executing such programs in ZK is the repeated evaluation of a CPU step, achieved by branching over the CPU's instruction set. This approach is general and covers traversal-execution of a program's control flow graph (CFG): here CPU instructions are straight-line program fragments (of various sizes) associated with the CFG nodes. This highlights the usefulness of ZK CPUs with a large number of instructions of varying sizes.
The rise of Autonomous Vehicles (AVs) brings with it the need for secure and privacy-preserving machine learning models. Federated Learning (FL) allows AVs to collaboratively train models while keeping raw data localized. However, traditional FL systems are vulnerable to security threats, including adversarial attacks, data breaches, and dependency on a central aggregator, which can be a single point of failure. To address these concerns, this paper introduces a peer-to-peer decentralized federated learning system that integrates lightweight blockchain technology and Binius Zero- Knowledge Proofs (ZKPs) to enhance security and privacy. In this system, Binius ZKPs ensure that model updates are cryptographically verified without exposing sensitive information, guaranteeing data confidentiality and integrity during the learning process. The lightweight blockchain framework secures the network by creating an immutable, decentralized record of all model updates, thus preventing tampering, fraud, or unauthorized modifications. This decentralized approach eliminates the need for a central aggregator, significantly enhancing system resilience to attacks and making it suitable for dynamic environments like AV networks. Additionally, the system's design includes Byzantine resilience, providing protection against adversarial nodes and ensuring that the global model aggregation process remains robust even in the presence of malicious actors. Extensive performance evaluations demonstrate that the system achieves low-latency, scalability, and efficient resource usage while maintaining strong security and privacy guarantees, making it an ideal solution for real-time federated learning in autonomous vehicle networks. The proposed framework not only ensures privacy but also fosters trust among participants in a fully decentralized environment.
With the demand of cryptocurrencies, threshold ECDSA recently regained popularity. So far, several methods have been proposed to construct threshold ECDSA, including the usage of OT and homomorphic encryptions (HE). Due to the mismatch between the plaintext space and the signature space, HE-based threshold ECDSA always requires zero-knowledge range proofs, such as Paillier and Joye-Libert (JL) encryptions. However, the overhead of range proofs constitutes a major portion of the total cost.
Ron Steinfeld, Amin Sakzad, Muhammed F. Esgin, Veronika Kuchta · 6 authors
We introduce the first candidate Lattice-based designated verifier (DV) zero knowledge sUccinct Non-interactive Argument (ZK-SNARG) protocol, named LUNA, with quasi-optimal proof length (quasi-linear in the security/privacy parameter). By simply relying on mildly stronger security assumptions, LUNA is also a candidate ZK-SNARK (i.e. argument of knowledge). LUNA achieves significant improvements in concrete proof sizes, reaching below 6 KB (compared to >32 KB in prior work) for 128-bit security/privacy level. To achieve our quasi-optimal succinct LUNA, we give a new regularity result for 'private' re-randomization of Module LWE (MLWE) samples using discrete Gaussian randomization vectors, also known as a lattice-based leftover hash lemma with leakage, which applies with a discrete Gaussian re-randomization parameter that is polynomial in the statistical privacy parameter (avoiding exponential smudging), and hides the coset of the re-randomization vector support set. Along the way, we derive bounds on the smoothing parameter of the intersection of short integer solution (SIS), gadget, and Gaussian perp module lattices over the power of 2 cyclotomic rings. We then introduce a new candidate linear-only homomorphic encryption scheme called Module Half-GSW (HGSW), and apply our regularity theorem to provide smudging-free circuit-private homomorphic linear operations for Module HGSW. Our implementation and experimental performance evaluation show that, for typical instance sizes, Module HGSW provides favourable performance for ZK-SNARG applications involving lightweight verifiers. It enables significantly (around 5x) shorter proof lengths while speeding up CRS generation and encryption time by 4-16x and speeding up decryption time by 4.3x, while incurring just 1.2-2x time overhead in linear homomorphic proof generation operations, compared to a Regev encryption used in prior work in the ZK-SNARG context. We believe our techniques are of independent interest and will find application in other privacy-preserving lattice-based protocols.
Michael L. Rosenberg, Tushar Mopuri, Hossein Hafezi, Ian Miers · 5 authors
Zero-knowledge Succinct Non-interactive ARguments of Knowledge (zkSNARKs) allow a prover to convince a verifier of the correct execution of a large computation in private and easily-verifiable manner.These properties make zkSNARKs a powerful tool for adding accountability, scalability, and privacy to numerous systems such as blockchains and verifiable key directories.Unfortunately, existing zkSNARKs are unable to scale to large computations due to time and space complexity requirements for the prover algorithm.As a result, they cannot handle real-world instances of the aforementioned applications.In this work, we introduce Hekaton, a zkSNARK that overcomes these barriers and can efficiently handle arbitrarily large computations.We construct Hekaton via a new "distribute-and-aggregate" framework that breaks up large computations into small chunks, proves these chunks in parallel in a distributed system, and then aggregates the resulting chunk proofs into a single succinct proof.Underlying this framework is a new technique for efficiently handling data that is shared between chunks that we believe could be of independent interest.We implement a distributed prover for Hekaton, and evaluate its performance on a compute cluster.Our experiments show that Hekaton achieves strong horizontal scalability (proving time decreases linearly as we increase the number of nodes in the cluster), and is able to prove large computations quickly: it can prove computations of size 2 35 gates in under an hour, which is much faster than prior work.Finally, we also apply Hekaton to two applications of realworld interest: proofs of batched insertion for a verifiable key directory and proving correctness of RAM computations.In both cases, Hekaton is able to scale to handle realistic workloads with better efficiency than prior work.
Space-efficient SNARKs aim to reduce the prover's space overhead which is one the main obstacles for deploying SNARKs in practice, as it can be prohibitively large (e.g., orders of magnitude larger than natively performing the computation). In this work, we propose Sparrow, a novel space-efficient zero-knowledge SNARK for data-parallel arithmetic circuits with two attractive features: (i) it is the first space-efficient scheme where, for a given field, the prover overhead increases with a multiplicative sublogarithmic factor as the circuit size increases, and (ii) compared to prior space-efficient SNARKs that work for arbitrary arithmetic circuits, it achieves prover space asymptotically smaller than the circuit size itself. Our key building block is a novel space-efficient sumcheck argument with improved prover time which may be of independent interest. Our experimental results for three use cases (arbitrary data parallel circuits, multiplication trees, batch SHA256 hashing) indicate Sparrow outperforms the prior state-of-the-art space-efficient SNARK for arithmetic circuits Gemini (Bootle et al., EUROCRYPT'22) by 3.2-28.7x in total prover space and 3.1-11.3x in prover time. We then use Sparrow to build zero-knowledge proofs of tree training and prediction, relying on its space efficiency to scale to large datasets and forests of multiple trees. Compared to a (non-space-efficient) optimal-time SNARK based on the GKR protocol, we observe prover space reduction of 16-240x for tree training while maintaining essentially the same prover and verifier times and proof size. Even more interestingly, our prover requires comparable space to natively perform the underlying computation. E.g., for a 400MB dataset, our prover only needs 1.4x more space than the native computation.
Interactive theorem provers (ITPs), such as Lean and Coq, can express formal proofs for a large category of theorems, from abstract math to software correctness. Consider Alice who has a Lean proof for some public statement T. Alice wants to convince the world that she has such a proof, without revealing the actual proof. Perhaps the proof shows that a secret program is correct or safe, but the proof itself might leak information about the program's source code. A natural way for Alice to proceed is to construct a succinct, zero-knowledge, non-interactive argument of knowledge (zkSNARK) to prove that she has a Lean proof for the statement T.
The hardness of lattice problems offers one of the most promising security foundations for quantum-safe cryptography. Basic schemes for public key encryption and digital signatures are already close to standardization at NIST and several other standardization bodies, and the research frontier has moved on to building primitives with more advanced privacy features. At the core of many such primitives are zero-knowledge proofs. In recent years, zero-knowledge proofs for (and using) lattice relations have seen a dramatic jump in efficiency and they currently provide arguably the shortest, and most computationally efficient, quantum-safe proofs for many scenarios. The main difficulty in using these proofs by non-experts (and experts!) is that they have a lot of moving parts and a lot of internal parameters depend on the particular instance that one is trying to prove.
Daniel Escudero, Antigoni Polychroniadou, Yifan Song, Chenkai Weng
In this work we study the efficiency of Zero-Knowledge (ZK) arguments of knowledge, particularly exploring Multi-Verifier ZK (MVZK) protocols as a midway point between Non-Interactive ZK and Designated-Verifier ZK, offering versatile applications across various domains. We introduce a new MVZK protocol designed for the preprocessing model, allowing any constant fraction of verifiers to be corrupted, potentially colluding with the prover. Our contributions include the first MVZK over rings. Unlike recent prior works on fields in the dishonest majority case, our protocol demonstrates communication complexity independent of the number of verifiers, contrasting the linear complexity of previous approaches. This key advancement ensures improved scalability and efficiency. We provide an end-to-end implementation of our protocol. The benchmark shows that it achieves a throughput of 1.47 million gates per second for 64 verifiers with 50% corruption, and 0.88 million gates per second with 75% corruption.
In the realm of healthcare analytics, preserving the privacy of sensitive data while enabling valuable insights poses a significant challenge, particularly given the increasing prevalence of data breaches and the sensitivity of personal health information. This paper presents a secure framework that addresses these concerns by integrating privacy-preserving parameters, zero-knowledge proofs (zk-SNARKs), blockchain technology, and a multi-tenant cloud environment. Through advanced cryptographic techniques, specifically zk-SNARKs, the framework ensures that healthcare records remain protected during analytics computations, without exposing raw data. The privacy-preserving analytics engine utilizes anonymized healthcare records and generates zk-SNARKs to validate computations. These proofs, integrated into a blockchain network, create a tamper-proof, transparent ledger that ensures secure healthcare transactions. This approach is critical in scenarios such as telemedicine, where secure data sharing and computation are paramount. By demonstrating its application in a telemedicine app, the framework highlights its practical significance in balancing data utility and privacy in healthcare analytics, providing a scalable and secure solution to a pressing problem.
Kasra Abbaszadeh, Christodoulos Pappas, Jonathan Katz, Dimitrios Papadopoulos
A zero-knowledge proof of training (zkPoT) enables a party to prove that they have correctly trained a committed model based on a committed dataset without revealing any additional information about the model or the dataset. An ideal zkPoT should offer provable security and privacy guarantees, succinct proof size and verifier runtime, and practical prover efficiency. In this work, we present Kaizen, a zkPoT targeted for deep neural networks (DNNs) that achieves all these goals at once. Our construction enables a prover to iteratively train their model via (mini-batch) gradient descent, where the number of iterations need not be fixed in advance; at the end of each iteration, the prover generates a commitment to the trained model parameters attached with a succinct zkPoT, attesting to the correctness of the executed iterations. The proof size and verifier time are independent of the number of iterations.
Jose Miguel Lara Rangel, Stefan Schoepf, J. Bennett Foster, David Krueger · 5 authors
Machine unlearning is gaining increasing attention as a way to remove adversarial data poisoning attacks from already trained models and to comply with privacy and AI regulations. The objective is to unlearn the effect of undesired data from a trained model while maintaining performance on the remaining data. This paper introduces HyperForget, a novel machine unlearning framework that leverages hypernetworks - neural networks that generate parameters for other networks - to dynamically sample models that lack knowledge of targeted data while preserving essential capabilities. Leveraging diffusion models, we implement two Diffusion HyperForget Networks and used them to sample unlearned models in Proof-of-Concept experiments. The unlearned models obtained zero accuracy on the forget set, while preserving good accuracy on the retain sets, highlighting the potential of HyperForget for dynamic targeted data removal and a promising direction for developing adaptive machine unlearning algorithms.
Blockchain technology has emerged as a transformative solution for securing distributed networks, offering decentralized and immutable data management. However, the resilience of blockchain systems faces challenges from various security threats, including double-spending, Sybil attacks, and vulnerabilities in smart contracts. This paper explores the effectiveness of various blockchain security protocols in enhancing the security and stability of distributed networks. The study provides a comprehensive review of cryptographic techniques, consensus algorithms, and privacy-enhancing technologies, such as Zero-Knowledge Proofs and Multi-Party Computation. Through a detailed analysis of case studies involving Bitcoin, Ethereum, and Hyperledger Fabric, the paper highlights the strengths and limitations of different security protocols. Additionally, the paper discusses the future direction of blockchain security, including the impact of emerging threats such as quantum computing on current security measures. The findings emphasize the need for ongoing innovation in security protocols to ensure the long-term resilience of blockchain networks. The paper concludes with recommendations for improving the security frameworks in both public and permissioned blockchains, with a focus on scalability, privacy, and resistance to emerging attacks.