Publication Library
Blockchain Security Risk Assessment in Quantum Era, Migration Strategies and Proactive Defense
Description: The emergence of quantum computing presents a formidable challenge to the security of blockchain systems. Traditional cryptographic algorithms, foundational to digital signatures, message encryption, and hashing functions, become vulnerable to the immense computational power of quantum computers. This paper conducts a thorough risk assessment of transitioning to quantum-resistant blockchains, comprehensively analyzing potential threats targeting vital blockchain components: the network, mining pools, transaction verification mechanisms, smart contracts, and user wallets. By elucidating the intricate challenges and strategic considerations inherent in transitioning to quantum-resistant algorithms, the paper evaluates risks and highlights obstacles in securing blockchain components with quantum-resistant cryptography. It offers a hybrid migration strategy to facilitate a smooth transition from classical to quantum-resistant cryptography. The analysis extends to prominent blockchains such as Bitcoin, Ethereum, Ripple, Litecoin, and Zcash, assessing vulnerable components, potential impacts, and associated STRIDE threats, thereby identifying areas susceptible to quantum attacks. Beyond analysis, the paper provides actionable guidance for designing secure and resilient blockchain ecosystems in the quantum computing era. Recognizing the looming threat of quantum computers, this research advocates for a proactive transition to quantum-resistant blockchain networks. It proposes a tailored security blueprint that strategically fortifies each component against the evolving landscape of quantum-induced cyber threats. Emphasizing the critical need for blockchain stakeholders to adopt proactive measures and implement quantum-resistant solutions, the paper underscores the importance of embracing these insights to navigate the complexities of the quantum era with resilience and confidence.
Created At: 22 January 2025
Updated At: 22 January 2025
Metamorphic Testing for Smart Contract Validation - A Case Study of Ethereum-Based Crowdfunding Contracts
Description: Blockchain smart contracts play a crucial role in automating and securing agreements in diverse domains such as finance, healthcare, and supply chains. Despite their critical applications, testing these contracts often receives less attention than their development, leaving significant risks due to the immutability of smart contracts post-deployment. A key challenge in the testing of smart contracts is the oracle problem, where the exact expected outcomes are not well defined, complicating systematic testing efforts. Metamorphic Testing (MT) addresses the oracle problem by using Metamorphic Relations (MRs) to validate smart contracts. MRs define how output should change relative to specific input modifications, determining whether the tests pass or fail. In this work, we apply MT to test an Ethereum-based crowdfunding smart contract, focusing on core functionalities such as state transitions and donation tracking. We identify a set of MRs tailored for smart contract testing and generate test cases for these MRs. To assess the effectiveness of this approach, we use the Vertigo mutation testing tool to create faulty versions of the smart contract. The experimental results show that our MRs detected 25.65% of the total mutants generated, with the most effective MRs achieving a mutant-killing rate of 89%. These results highlight the utility of MT to ensure the reliability and quality of blockchain-based smart contracts.
Created At: 22 January 2025
Updated At: 22 January 2025
A mixture transition distribution approach to portfolio optimization
Description: Understanding the dependencies among financial assets is critical for portfolio optimization. Traditional approaches based on correlation networks often fail to capture the nonlinear and directional relationships that exist in financial markets. In this study, we construct directed and weighted financial networks using the Mixture Transition Distribution (MTD) model, offering a richer representation of asset interdependencies. We apply local assortativity measures--metrics that evaluate how assets connect based on similarities or differences--to guide portfolio selection and allocation. Using data from the Dow Jones 30, Euro Stoxx 50, and FTSE 100 indices constituents, we show that portfolios optimized with network-based assortativity measures consistently outperform the classical mean-variance framework. Notably, modalities in which assets with differing characteristics connect enhance diversification and improve Sharpe ratios. The directed nature of MTD-based networks effectively captures complex relationships, yielding portfolios with superior risk-adjusted returns. Our findings highlight the utility of network-based methodologies in financial decision-making, demonstrating their ability to refine portfolio optimization strategies. This work thus underscores the potential of leveraging advanced financial networks to achieve enhanced performance, offering valuable insights for practitioners and setting a foundation for future research.
Created At: 22 January 2025
Updated At: 22 January 2025
LLM-Powered Multi-Agent System for Automated Crypto Portfolio Management
Description: Cryptocurrency investment is inherently difficult due to its shorter history compared to traditional assets, the need to integrate vast amounts of data from various modalities, and the requirement for complex reasoning. While deep learning approaches have been applied to address these challenges, their black-box nature raises concerns about trust and explainability. Recently, large language models (LLMs) have shown promise in financial applications due to their ability to understand multi-modal data and generate explainable decisions. However, single LLM faces limitations in complex, comprehensive tasks such as asset investment. These limitations are even more pronounced in cryptocurrency investment, where LLMs have less domain-specific knowledge in their training corpora. To overcome these challenges, we propose an explainable, multi-modal, multi-agent framework for cryptocurrency investment. Our framework uses specialized agents that collaborate within and across teams to handle subtasks such as data analysis, literature integration, and investment decision-making for the top 30 cryptocurrencies by market capitalization. The expert training module fine-tunes agents using multi-modal historical data and professional investment literature, while the multi-agent investment module employs real-time data to make informed cryptocurrency investment decisions. Unique intrateam and interteam collaboration mechanisms enhance prediction accuracy by adjusting final predictions based on confidence levels within agent teams and facilitating information sharing between teams. Empirical evaluation using data from November 2023 to September 2024 demonstrates that our framework outperforms single-agent models and market benchmarks in classification, asset pricing, portfolio, and explainability performance.
Created At: 22 January 2025
Updated At: 22 January 2025
Towards Human-Guided, Data-Centric LLM Co-Pilots
Description: Machine learning (ML) has the potential to revolutionize healthcare, but its adoption is often hindered by the disconnect between the needs of domain experts and translating these needs into robust and valid ML tools. Despite recent advances in LLM-based co-pilots to democratize ML for non-technical domain experts, these systems remain predominantly focused on model-centric aspects while overlooking critical data-centric challenges. This limitation is problematic in complex real-world settings where raw data often contains complex issues, such as missing values, label noise, and domain-specific nuances requiring tailored handling. To address this we introduce CliMB-DC, a human-guided, data-centric framework for LLM co-pilots that combines advanced data-centric tools with LLM-driven reasoning to enable robust, context-aware data processing. At its core, CliMB-DC introduces a novel, multi-agent reasoning system that combines a strategic coordinator for dynamic planning and adaptation with a specialized worker agent for precise execution. Domain expertise is then systematically incorporated to guide the reasoning process using a human-in-the-loop approach. To guide development, we formalize a taxonomy of key data-centric challenges that co-pilots must address. Thereafter, to address the dimensions of the taxonomy, we integrate state-of-the-art data-centric tools into an extensible, open-source architecture, facilitating the addition of new tools from the research community. Empirically, using real-world healthcare datasets we demonstrate CliMB-DC's ability to transform uncurated datasets into ML-ready formats, significantly outperforming existing co-pilot baselines for handling data-centric challenges. CliMB-DC promises to empower domain experts from diverse domains -- healthcare, finance, social sciences and more -- to actively participate in driving real-world impact using ML.
Created At: 22 January 2025
Updated At: 22 January 2025