arXiv subjects

Tomaso Aste

Publications and source records attributed to Tomaso Aste.

At least 19 recordsLinked to original sources

arXiv2026

When Agents Coordinate: Measuring Coordination in Multi-Agent AI Coding

We study how teams of AI coding agents coordinate while solving programming tasks. Current evaluations usually report whether the agents complete the task and how much the run costs, leaving the coordination inside the team largely unmeasured. We introduce an instrument to measure this coordination. Each run is represented as a temporal network in which agents and files are nodes, and messages, file writes, and file reads are timestamped directed edges with an associated cost. We apply this instrument to 1902 runs, each evaluated with a fixed test suite, across configurations that vary the team size, the team structure, and the file policy. The resulting networks show how coordination changes as teams grow and as the work changes. Direct messaging initially increases close to quadratically with the number of agents, with much of this growth coming from an early round of introductions. As the teams grow further, this increase levels off in the largest teams we study, where agents increasingly communicate through broadcast messages. The task also shapes the network that emerges. Work built around a shared specification produces dense, highly connected teams, while pipeline tasks produce sparse networks organised around local interfaces. Shared files can replace repeated 1-to-1 communication, cutting output tokens by about 42% at eight agents on message-heavy work, while adding overhead when files already carry the coordination. Naming one agent as coordinator creates no communication hub and provides no reliable improvement in success. We also observe an unprompted tendency for agents to seek out hidden grading material. We repeat the key experimental conditions in a sealed environment, replacing the hidden material with marked placeholder files. Across 244 additional runs, agents still reach for it in four fifths of runs, while the coordinator and file-channel findings reproduce.

cs.AI↗

arXiv2026

Dependence-Informed Sparse Neural Architecture for Stock Return Prediction

Using neural networks for stock return prediction typically requires choices about depth and hidden-layer width that are difficult to connect to financial interpretation. We study an alternative: estimate dependence among firm characteristics with a Maximally Filtered Clique Forest (MFCF), then map its clique structure to a Homological Neural Network (HNN). The MFCF maximum clique size K is the only parameter controlling architectural complexity, and it has a clear graphical meaning: it bounds the number of characteristics in each maximal clique and hence the highest interaction order the network can represent. The filtered graph then fixes the neural network's depth, layer widths, and sparse connections before training, in place of a separately chosen depth and width sequence. We apply two HNN variants to annual out-of-sample forecasts of U.S. stock excess returns from 1987 to 2016 using 94 firm characteristics. The HNN models match a three-hidden-layer benchmark on pooled predictive accuracy, rank the cross-section more accurately, and use roughly 80 times fewer parameters than a fully connected network with the same induced layer widths. Two structural ablations indicate that both the sparse connectivity and the estimated grouping of characteristics contribute to the ranking advantage, and both effects remain significant after correcting for multiple testing. These findings show that HNNs offer a practical and interpretable way to incorporate estimated dependence among firm characteristics into neural architecture design.

q-fin.ST↗

arXiv2026

Smooth Neural Point Processes via B-Splines

Temporal point processes (TPPs) provide a general and flexible framework for modeling sequences of events in continuous time. Neural networks have been successfully employed to model TPPs in a highly expressive and data-driven way. Neural TPPs are typically trained via Maximum Likelihood Estimation (MLE) by minimizing the negative log-likelihood (NLL), which depends on both the conditional intensity function (CIF) and its integral over time, the compensator. Recent neural TPP approaches enable exact evaluation of the NLL without numerical integration. However, these methods typically model the compensator rather than the CIF directly, impose constraints on the neural network architecture, and are computationally expensive during training, as event contributions to the NLL are evaluated sequentially rather than in parallel. In this work, we propose a novel neural TPP model that directly parametrizes the CIF as a non-negative combination of B-spline basis functions, whose coefficients are predicted by a neural network. This formulation enables exact evaluation of the NLL, preserves full flexibility in the neural architecture, allows efficient parallelization during training, and naturally supports CIF smoothness regularization through the integrated squared second derivative. Experiments on both synthetic and real-world datasets show improved computational efficiency and predictive accuracy compared to the reference neural TPP baseline.

cs.LG↗

arXiv2026

Compositional Sparsity as an Inductive Bias for Neural Architecture Design

Identifying the structural priors that enable Deep Neural Networks (DNNs) to overcome the curse of dimensionality is a fundamental challenge in machine learning theory. Existing literature suggests that effective high-dimensional learning is driven by compositional sparsity, where target functions decompose into constituents supported on low-dimensional variable subsets. To investigate this hypothesis, we combine Information Filtering Networks (IFNs), which extract sparse dependency structures via constrained information maximisation, with Homological Neural Networks (HNNs), which map the inferred topology into fixed-wiring sparse neural graphs. We formalise the design principles underlying this construction and present an interpretable pipeline in which abstraction emerges through hierarchical composition. HNNs are orders of magnitude sparser than standard DNNs and require only minimal hyperparameter tuning. On synthetic tasks with known sparse hierarchies, HNNs recover the underlying compositional structure and remain stable in regimes where dense alternatives degrade as dimensionality increases. Across a broad suite of real-world datasets, HNNs consistently match or outperform dense baselines while using far fewer parameters, exhibiting lower variance and showing reduced sensitivity to hyperparameters.

cs.LG↗

arXiv2026

Graph Regularized PCA

Multivariate data often exhibit complex dependencies that violate the assumption of isotropic residual noise. For such cases, we introduce Graph Regularized PCA (GR-PCA). It is a graph-based regularization of PCA that incorporates the dependency structure of the data features by learning a sparse precision graph and biasing loadings toward the low-frequency Fourier modes of the corresponding graph Laplacian. Consequently, high-frequency signals are suppressed, while graph-coherent low-frequency ones are preserved, yielding interpretable principal components aligned with conditional relationships. We evaluate GR-PCA on synthetic data spanning diverse graph topologies, signal-to-noise ratios, and sparsity levels. Compared to mainstream alternatives, it concentrates variance on the intended support, produces loadings with lower graph-Laplacian energy, and remains competitive in out-of-sample reconstruction. When high-frequency signals are present, the graph Laplacian penalty prevents overfitting, reducing the reconstruction accuracy but improving structural fidelity. The advantage over PCA is most pronounced when high-frequency signals are graph-correlated, whereas PCA remains competitive when such signals are nearly rotationally invariant. The procedure is simple to implement, modular with respect to the precision estimator, and scalable, providing a practical route to structure-aware dimensionality reduction that improves structural fidelity without sacrificing predictive performance.

cs.LG↗

arXiv2025

Cryptocurrencies in the Balance Sheet: Insights from (Micro)Strategy -- Bitcoin Interactions

This paper investigates the evolving link between cryptocurrency and equity markets in the context of the recent wave of corporate Bitcoin (BTC) treasury strategies. We assemble a dataset of 39 publicly listed firms holding BTC, from their first acquisition through April 2025. Using daily logarithmic returns, we first document significant positive co-movements via Pearson correlations and single factor model regressions, discovering an average BTC beta of 0.62, and isolating 12 companies, including Strategy (formerly MicroStrategy, MSTR), exhibiting a beta exceeding 1. We then classify firms into three groups reflecting their exposure to BTC, liquidity, and return co-movements. We use transfer entropy (TE) to capture the direction of information flow over time. Transfer entropy analysis consistently identifies BTC as the dominant information driver, with brief, announcement-driven feedback from stocks to BTC during major financial events. Our results highlight the critical need for dynamic hedging ratios that adapt to shifting information flows. These findings provide important insights for investors and managers regarding risk management and portfolio diversification in a period of growing integration of digital assets into corporate treasuries.

q-fin.GN↗

arXiv2025

Information Filtering Networks: Theoretical Foundations, Generative Methodologies, and Real-World Applications

Information Filtering Networks (IFNs) provide a powerful framework for modeling complex systems through globally sparse yet locally dense and interpretable structures that capture multivariate dependencies. This review offers a comprehensive account of IFNs, covering their theoretical foundations, construction methodologies, and diverse applications. Tracing their origins from early network-based models to advanced formulations such as the Triangulated Maximally Filtered Graph (TMFG) and the Maximally Filtered Clique Forest (MFCF), the paper highlights how IFNs address key challenges in high-dimensional data-driven modeling. IFNs and their construction methodologies are intrinsically higher-order networks that generate simplicial complexes-structures that are only now becoming popular in the broader literature. Applications span fields including finance, biology, psychology, and artificial intelligence, where IFNs improve interpretability, computational efficiency, and predictive performance. Special attention is given to their role in graphical modeling, where IFNs enable the estimation of sparse inverse covariance matrices with greater accuracy and scalability than traditional approaches like Graphical LASSO. Finally, the review discusses recent developments that integrate IFNs with machine learning and deep learning, underscoring their potential not only to bridge classical network theory with contemporary data-driven paradigms, but also to shape the architectures of deep learning models themselves.

cs.LG↗

arXiv2024

Granger Causality Detection with Kolmogorov-Arnold Networks

Discovering causal relationships in time series data is central in many scientific areas, ranging from economics to climate science. Granger causality is a powerful tool for causality detection. However, its original formulation is limited by its linear form and only recently nonlinear machine-learning generalizations have been introduced. This study contributes to the definition of neural Granger causality models by investigating the application of Kolmogorov-Arnold networks (KANs) in Granger causality detection and comparing their capabilities against multilayer perceptrons (MLP). In this work, we develop a framework called Granger Causality KAN (GC-KAN) along with a tailored training approach designed specifically for Granger causality detection. We test this framework on both Vector Autoregressive (VAR) models and chaotic Lorenz-96 systems, analysing the ability of KANs to sparsify input features by identifying Granger causal relationships, providing a concise yet accurate model for Granger causality detection. Our findings show the potential of KANs to outperform MLPs in discerning interpretable Granger causal relationships, particularly for the ability of identifying sparse Granger causality patterns in high-dimensional settings, and more generally, the potential of AI in causality discovery for the dynamical laws in physical systems.

cs.LG↗

arXiv2024

HLOB -- Information Persistence and Structure in Limit Order Books

We introduce a novel large-scale deep learning model for Limit Order Book mid-price changes forecasting, and we name it `HLOB'. This architecture (i) exploits the information encoded by an Information Filtering Network, namely the Triangulated Maximally Filtered Graph, to unveil deeper and non-trivial dependency structures among volume levels; and (ii) guarantees deterministic design choices to handle the complexity of the underlying system by drawing inspiration from the groundbreaking class of Homological Convolutional Neural Networks. We test our model against 9 state-of-the-art deep learning alternatives on 3 real-world Limit Order Book datasets, each including 15 stocks traded on the NASDAQ exchange, and we systematically characterize the scenarios where HLOB outperforms state-of-the-art architectures. Our approach sheds new light on the spatial distribution of information in Limit Order Books and on its degradation over increasing prediction horizons, narrowing the gap between microstructural modeling and deep learning-based forecasting in high-frequency financial markets.

q-fin.TR↗

arXiv2024

Deep Limit Order Book Forecasting

We exploit cutting-edge deep learning methodologies to explore the predictability of high-frequency Limit Order Book mid-price changes for a heterogeneous set of stocks traded on the NASDAQ exchange. In so doing, we release `LOBFrame', an open-source code base to efficiently process large-scale Limit Order Book data and quantitatively assess state-of-the-art deep learning models' forecasting capabilities. Our results are twofold. We demonstrate that the stocks' microstructural characteristics influence the efficacy of deep learning methods and that their high forecasting power does not necessarily correspond to actionable trading signals. We argue that traditional machine learning metrics fail to adequately assess the quality of forecasts in the Limit Order Book context. As an alternative, we propose an innovative operational framework that evaluates predictions' practicality by focusing on the probability of accurately forecasting complete transactions. This work offers academics and practitioners an avenue to make informed and robust decisions on the application of deep learning techniques, their scope and limitations, effectively exploiting emergent statistical properties of the Limit Order Book.

q-fin.TR↗

arXiv2024

Retail Central Bank Digital Currency: Motivations, Opportunities, and Mistakes

Nations around the world are conducting research into the design of central bank digital currency (CBDC), a new, digital form of money that would be issued by central banks alongside cash and central bank reserves. Retail CBDC would be used by individuals and businesses as form of money suitable for routine commerce. An important motivating factor in the development of retail CBDC is the decline of the popularity of central bank money for retail purchases and the increasing use of digital money created by the private sector for such purposes. The debate about how retail CBDC would be designed and implemented has led to many proposals, which have sparked considerable debate about business models, regulatory frameworks, and the socio-technical role of money in general. Here, we present a critical analysis of the existing proposals. We examine their motivations and themes, as well as their underlying assumptions. We also offer a reflection of the opportunity that retail CBDC represents and suggest a way forward in furtherance of the public interest.

cs.CY↗

arXiv2023

Topological Portfolio Selection and Optimization

Modern portfolio optimization is centered around creating a low-risk portfolio with extensive asset diversification. Following the seminal work of Markowitz, optimal asset allocation can be computed using a constrained optimization model based on empirical covariance. However, covariance is typically estimated from historical lookback observations, and it is prone to noise and may inadequately represent future market behavior. As a remedy, information filtering networks from network science can be used to mitigate the noise in empirical covariance estimation, and therefore, can bring added value to the portfolio construction process. In this paper, we propose the use of the Statistically Robust Information Filtering Network (SR-IFN) which leverages the bootstrapping techniques to eliminate unnecessary edges during the network formation and enhances the network's noise reduction capability further. We apply SR-IFN to index component stock pools in the US, UK, and China to assess its effectiveness. The SR-IFN network is partially disconnected with isolated nodes representing lesser-correlated assets, facilitating the selection of peripheral, diversified and higher-performing portfolios. Further optimization of performance can be achieved by inversely proportioning asset weights to their centrality based on the resultant network.

q-fin.PM↗

arXiv2023

Unraveling the Enigma of Double Descent: An In-depth Analysis through the Lens of Learned Feature Space

Double descent presents a counter-intuitive aspect within the machine learning domain, and researchers have observed its manifestation in various models and tasks. While some theoretical explanations have been proposed for this phenomenon in specific contexts, an accepted theory to account for its occurrence in deep learning remains yet to be established. In this study, we revisit the phenomenon of double descent and demonstrate that its occurrence is strongly influenced by the presence of noisy data. Through conducting a comprehensive analysis of the feature space of learned representations, we unveil that double descent arises in imperfect models trained with noisy data. We argue that double descent is a consequence of the model first learning the noisy data until interpolation and then adding implicit regularization via over-parameterization acquiring therefore capability to separate the information from the noise.

cs.LG↗

arXiv2023

Homological Convolutional Neural Networks

Deep learning methods have demonstrated outstanding performances on classification and regression tasks on homogeneous data types (e.g., image, audio, and text data). However, tabular data still pose a challenge, with classic machine learning approaches being often computationally cheaper and equally effective than increasingly complex deep learning architectures. The challenge arises from the fact that, in tabular data, the correlation among features is weaker than the one from spatial or semantic relationships in images or natural language, and the dependency structures need to be modeled without any prior information. In this work, we propose a novel deep learning architecture that exploits the data structural organization through topologically constrained network representations to gain relational information from sparse tabular inputs. The resulting model leverages the power of convolution and is centered on a limited number of concepts from network topology to guarantee: (i) a data-centric and deterministic building pipeline; (ii) a high level of interpretability over the inference process; and (iii) an adequate room for scalability. We test our model on 18 benchmark datasets against 5 classic machine learning and 3 deep learning models, demonstrating that our approach reaches state-of-the-art performances on these challenging datasets. The code to reproduce all our experiments is provided at https://github.com/FinancialComputingUCL/HomologicalCNN.

cs.LG↗

arXiv2023

ESG Reputation Risk Matters: An Event Study Based on Social Media Data

We investigate the response of shareholders to Environmental, Social, and Governance-related reputational risk (ESG-risk), focusing exclusively on the impact of social media. Using a dataset of 114 million tweets about firms listed on the S&P100 index between 2016 and 2022, we extract conversations discussing ESG matters. In an event study design, we define events as unusual spikes in message posting activity linked to ESG-risk, and we then examine the corresponding changes in the returns of related assets. By focusing on social media, we gain insight into public opinion and investor sentiment, an aspect not captured through ESG controversies news alone. To the best of our knowledge, our approach is the first to distinctly separate the reputational impact on social media from the physical costs associated with negative ESG controversy news. Our results show that the occurrence of an ESG-risk event leads to a statistically significant average reduction of 0.29% in abnormal returns. Furthermore, our study suggests this effect is predominantly driven by Social and Governance categories, along with the "Environmental Opportunities" subcategory. Our research highlights the considerable impact of social media on financial markets, particularly in shaping shareholders' perception of ESG reputation. We formulate several policy implications based on our findings.

econ.GN↗

arXiv2023

Homological Neural Networks: A Sparse Architecture for Multivariate Complexity

The rapid progress of Artificial Intelligence research came with the development of increasingly complex deep learning models, leading to growing challenges in terms of computational complexity, energy efficiency and interpretability. In this study, we apply advanced network-based information filtering techniques to design a novel deep neural network unit characterized by a sparse higher-order graphical architecture built over the homological structure of underlying data. We demonstrate its effectiveness in two application domains which are traditionally challenging for deep learning: tabular data and time series regression problems. Results demonstrate the advantages of this novel design which can tie or overcome the results of state-of-the-art machine learning and deep learning models using only a fraction of parameters.

cs.LG↗

arXiv2023

FTX's downfall and Binance's consolidation: The fragility of centralised digital finance

This paper investigates the causes of the FTX digital currency exchange's failure in November 2022. We identify the collapse of the Terra-Luna ecosystem as the pivotal event that triggered a significant decrease in the exchange's liquidity. Analysing on-chain data, we report that FTX heavily relied on leveraging and misusing its native token, FTT, and we show how this behaviour exacerbated the company's fragile financial situation. To gain further insights into the downfall, we study evolutionary dependency structures of 199 cryptocurrencies on an hourly basis, and we investigate public trades at the time of the events. Results suggest that the collapse was actively accelerated by Binance tweets causing a systemic reaction in the cryptocurrency market. Finally, identifying the actors who mostly benefited from the FTX's collapse and highlighting a generalised trend toward centralisation in the crypto space, we emphasise the importance of genuinely decentralised finance for a transparent, future digital economy.

q-fin.GN↗

arXiv2023

Topological Feature Selection

In this paper, we introduce a novel unsupervised, graph-based filter feature selection technique which exploits the power of topologically constrained network representations. We model dependency structures among features using a family of chordal graphs (the Triangulated Maximally Filtered Graph), and we maximise the likelihood of features' relevance by studying their relative position inside the network. Such an approach presents three aspects that are particularly satisfactory compared to its alternatives: (i) it is highly tunable and easily adaptable to the nature of input data; (ii) it is fully explainable, maintaining, at the same time, a remarkable level of simplicity; (iii) it is computationally cheaper compared to its alternatives. We test our algorithm on 16 benchmark datasets from different applicative domains showing that it outperforms or matches the current state-of-the-art under heterogeneous evaluation conditions.

cs.LG↗