arXiv subjects

Vaiva Vasiliauskaite

Publications and source records attributed to Vaiva Vasiliauskaite.

11 recordsLinked to original sources

arXiv2025

Directed Cycles as Higher-Order Units of Information Processing in Complex Networks

Directed cycles form the fundamental motifs in natural, social and artificial networks, yet their distinct computational roles remain under-explored, particularly in the context of higher-order structure and function. In this work, we investigate how two types of directed cycles - feedforward and feedback - can act as higher-order structures to facilitate the flow and integration of information in sparse random networks, and how these roles depend on the environment of the cycles. Using information-theoretic measures, we show that network size, sparsity and relative directionality critically impact the information-processing capacities of directed cycles. In a network with no-preferred global direction, a feedforward cycle enables greater information flow and a feedback cycle allows for increased information integration. The relative direction of a feedforward cycle as well as the structural incoherence it induces, determines its capacity to generate higher-order behaviour. Finally, we demonstrate that introducing feedback loops into otherwise feedforward architectures increases the diversity of network activity patterns. These findings suggest that directed cycles serve as computational motifs with local information processing capabilities that depend on the structure they are embedded. Using directed cycles, we highlight the interdependence between higher-order structures and the higher-order order behaviour they can induce in the network dynamics.

cond-mat.stat-mech↗

arXiv2024

Stretched and measured neural predictions of complex network dynamics

Differential equations are a ubiquitous tool to study dynamics, ranging from physical systems to complex systems, where a large number of agents interact through a graph with non-trivial topological features. Data-driven approximations of differential equations present a promising alternative to traditional methods for uncovering a model of dynamical systems, especially in complex systems that lack explicit first principles. A recently employed machine learning tool for studying dynamics is neural networks, which can be used for data-driven solution finding or discovery of differential equations. Specifically for the latter task, however, deploying deep learning models in unfamiliar settings - such as predicting dynamics in unobserved state space regions or on novel graphs - can lead to spurious results. Focusing on complex systems whose dynamics are described with a system of first-order differential equations coupled through a graph, we show that extending the model's generalizability beyond traditional statistical learning theory limits is feasible. However, achieving this advanced level of generalization requires neural network models to conform to fundamental assumptions about the dynamical model. Additionally, we propose a statistical significance test to assess prediction quality during inference, enabling the identification of a neural network's confidence level in its predictions.

cond-mat.stat-mech↗

arXiv2022

Ask "Who", Not "What": Bitcoin Volatility Forecasting with Twitter Data

Understanding the variations in trading price (volatility), and its response to exogenous information, is a well-researched topic in finance. In this study, we focus on finding stable and accurate volatility predictors for a relatively new asset class of cryptocurrencies, in particular Bitcoin, using deep learning representations of public social media data obtained from Twitter. For our experiments, we extracted semantic information and user statistics from over 30 million Bitcoin-related tweets, in conjunction with 15-minute frequency price data over a horizon of 144 days. Using this data, we built several deep learning architectures that utilized different combinations of the gathered information. For each model, we conducted ablation studies to assess the influence of different components and feature sets over the prediction accuracy. We found statistical evidences for the hypotheses that: (i) temporal convolutional networks perform significantly better than both classical autoregressive models and other deep learning-based architectures in the literature, and (ii) tweet author meta-information, even detached from the tweet itself, is a better predictor of volatility than the semantic content and tweet volume statistics. We demonstrate how different information sets gathered from social media can be utilized in different architectures and how they affect the prediction results. As an additional contribution, we make our dataset public for future research.

q-fin.ST↗

arXiv2022

Topic Community Based Temporal Expertise for Question Routing

Question Routing in Community-based Question Answering websites aims at recommending newly posted questions to potential users who are most likely to provide "accepted answers". Most of the existing approaches predict users' expertise based on their past question answering behavior and the content of new questions. However, these approaches suffer from challenges in three aspects: 1) sparsity of users' past records results in lack of personalized recommendation that at times does not match users' interest or domain expertise, 2) modeling based on all questions and answers content makes periodic updates computationally expensive, and 3) while CQA sites are highly dynamic, they are mostly considered as static. This paper proposes a novel approach to QR that addresses the above challenges. It is based on dynamic modeling of users' activity on topic communities. Experimental results on three real-world datasets demonstrate that the proposed model significantly outperforms competitive baseline models

cs.SI↗

arXiv2021

Information dynamics of price and liquidity around the 2017 Bitcoin markets crash

We study the information dynamics between the largest Bitcoin exchange markets during the bubble in 2017-2018. By analysing high-frequency market-microstructure observables with different information theoretic measures for dynamical systems, we find temporal changes in information sharing across markets. In particular, we study the time-varying components of predictability, memory, and synchronous coupling, measured by transfer entropy, active information storage, and multi-information. By comparing these empirical findings with several models we argue that some results could relate to intra-market and inter-market regime shifts, and changes in direction of information flow between different market observables.

q-fin.ST↗

arXiv2021

Cycle Analysis of Directed Acyclic Graphs

In this paper, we employ the decomposition of a directed network as an undirected graph plus its associated node metadata to characterise the cyclic structure found in directed networks by finding a Minimal Cycle Basis of the undirected graph and augment its components with direction information. We show that only four classes of directed cycles exist, and that they can be fully distinguished by the organisation and number of source-sink node pairs and their antichain structure. We are particularly interested in Directed Acyclic Graphs and introduce a set of metrics that characterise the Minimal Cycle Basis using the Directed Acyclic Graphs metadata information. In particular, we numerically show that Transitive Reduction stabilises the properties of Minimal Cycle Bases measured by the metrics we introduced while retaining key properties of the Directed Acyclic Graph. This makes the metrics consistent characterisation of Directed Acyclic Graphs and the systems they represent. We measure the characteristics of the Minimal Cycle Bases of four models of Transitively Reduced Directed Acyclic Graphs and show that the metrics introduced are able to distinguish the models and are sensitive to their generating mechanisms.

cs.SI↗

arXiv2021

Some Challenges in Monitoring Epidemics

Epidemic models often reflect characteristic features of infectious spreading processes by coupled non-linear differential equations considering different states of health (such as Susceptible, Infected, or Recovered). This compartmental modeling approach, however, delivers an incomplete picture of the dynamics of epidemics, as it neglects stochastic and network effects, and also the role of the measurement process, on which the estimation of epidemiological parameters and incidence values relies. In order to study the related issues, we extend established epidemiological spreading models with a model of the measurement (i.e. testing) process, considering the problems of false positives and false negatives as well as biased sampling. Studying a model-generated ground truth in conjunction with simulated observation processes (virtual measurements) allows one to gain insights into the limitations of purely data-driven methods to assess the epidemic situation. We conclude that epidemic monitoring, simulation, and forecasting are wicked problems, as applying a conventional data-driven approach to a complex system with non-linear dynamics, network effects, and uncertainty can be misleading. Nevertheless, some of the errors can be corrected for, using scientific knowledge of the spreading dynamics and the measurement process. We conclude that such corrections should generally be part of epidemic monitoring, modeling, and forecasting efforts.

physics.soc-ph↗

arXiv2020

Longest Path in the Price Model

The Price model, the directed version of the Barabási-Albert model, produces a growing directed acyclic graph. We look at variants of the model in which directed edges are added to the new vertex in one of two ways: using cumulative advantage (preferential attachment) choosing vertices in proportion to their degree, or with random attachment in which vertices are chosen uniformly at random. In such networks, the longest path is well defined and in some cases is known to be a better approximation to geodesics than the shortest path. We define a reverse greedy path and show both analytically and numerically that this scales with the logarithm of the size of the network with a coefficient given by the number of edges added using random attachment. This is a lower bound on the length of the longest path to any given vertex and we show numerically that the longest path also scales with the logarithm of the size of the network but with a larger coefficient that has some weak dependence on the parameters of the model.

physics.soc-ph↗

arXiv2020

Understanding complexity via network theory: a gentle introduction

Network theory provides tools which are particularly appropriate for assessing the complex interdependencies that characterise our modern connected world. This article presents an introduction to network theory, in a way that doesn't require a strong mathematical background. We explore how network theory unveils commonalities in the interdependency profiles of various systems, ranging from biological, to social, and artistic domains. Our aim is to enable an intuitive understanding while conveying the fundamental principles and aims of complexity science. Additionally, various network-theoretic tools are discussed, and numerous references for more advanced materials are provided.

physics.soc-ph↗

arXiv2020

Making Communities Show Respect for Order

In this work we give a community detection algorithm in which the communities both respects the intrinsic order of a directed acyclic graph and also finds similar nodes. We take inspiration from classic similarity measures of bibliometrics, used to assess how similar two publications are, based on their relative citation patterns. We study the algorithm's performance and antichain properties in artificial models and in real networks, such as citation graphs and food webs. We show how well this partitioning algorithm distinguishes and groups together nodes of the same origin (in a citation network, the origin is a topic or a research field). We make the comparison between our partitioning algorithm and standard hierarchical layering tools as well as community detection methods. We show that our algorithm produces different communities from standard layering algorithms.

physics.soc-ph↗

arXiv2019

Social Success of Perfumes

We study data on perfumes and their odour descriptors - notes - to understand how note compositions, called accords, influence successful fragrance formulas. We obtain accords which tend to be present in perfumes that receive significantly more customer ratings. Our findings show that the most popular notes and the most over-represented accords are different to those that have the strongest effect to the perfume ratings. We also used network centrality to understand which notes have the highest potential to enhance note compositions. We find that large degree notes, such as musk and vanilla as well as generically-named notes, e.g. floral notes, are amongst the notes that enhance accords the most. This work presents a framework which would be a timely tool for perfumers to explore a multidimensional space of scent compositions.

physics.soc-ph↗