arXiv subjects

James Bagrow

Publications and source records attributed to James Bagrow.

15 recordsLinked to original sources

arXiv2026

KANs need curvature: penalties for compositional smoothness

Kolmogorov-Arnold networks (KANs) offer a potent combination of accuracy and interpretability, thanks to their compositions of learnable univariate activation functions. However, the activations of well-fitting KANs tend to exhibit pathologically high-curvature oscillations, making them difficult to interpret, and standard regularization penalties do not prevent this. Here we derive a basis-agnostic curvature penalty and show that penalized models can maintain accuracy while achieving substantially smoother activations. Accounting for how function composition shapes curvature, we prove an upper bound on the full model's curvature relative to the curvature penalty, and use this to motivate richer forms of penalties. Scientific machine learning is increasingly bottlenecked by the trade-off between accuracy and interpretability. Results such as ours that improve interpretability without sacrificing accuracy will further strengthen KANs as a practical tool for both prediction and insight.

cs.LG↗

arXiv2025

Optimized Architectures for Kolmogorov-Arnold Networks

Efforts to improve Kolmogorov--Arnold networks (KANs) with architectural enhancements have been stymied by the complexity those enhancements bring, undermining the interpretability that makes KANs attractive in the first place. Here we study overprovisioned architectures combined with sparsification, deep supervision, and depth selection, to learn compact, interpretable KANs without sacrificing accuracy. Crucially, we focus on differentiable mechanisms under a principled minimum description length objective, jointly optimizing activations, structure, and depth end-to-end. Experiments across function approximation benchmarks, dynamical systems forecasting, and real-world prediction tasks demonstrate that sparsification alone is insufficient, but the combination with depth selection achieves competitive or superior accuracy while discovering substantially smaller models. The result is a principled path toward models that are both more expressive and more interpretable, addressing a key tension in scientific machine learning.

cs.LG↗

arXiv2025

Softly Symbolifying Kolmogorov-Arnold Networks

Kolmogorov-Arnold Networks (KANs) offer a promising path toward interpretable machine learning: their learnable activations can be studied individually, while collectively fitting complex data accurately. In practice, however, trained activations often lack symbolic fidelity, learning pathological decompositions with no meaningful correspondence to interpretable forms. We propose Softly Symbolified Kolmogorov-Arnold Networks (S2KAN), which integrate symbolic primitives directly into training. Each activation draws from a dictionary of symbolic and dense terms, with learnable gates that sparsify the representation. Crucially, this sparsification is differentiable, enabling end-to-end optimization, and is guided by a principled Minimum Description Length objective. When symbolic terms suffice, S2KAN discovers interpretable forms; when they do not, it gracefully degrades to dense splines. We demonstrate competitive or superior accuracy with substantially smaller models across symbolic benchmarks, dynamical systems forecasting, and real-world prediction tasks, and observe evidence of emergent self-sparsification even without regularization pressure.

cs.LG↗

arXiv2025

Vertex addition to a ball graph with application to reliability and area coverage in autonomous swarms

A unit ball graph consists of a set of vertices, labeled by points in Euclidean space, and edges joining all pairs of points within distance 1. These geometric graphs are used to model a variety of spatial networks, including communication networks between agents in an autonomous swarm. In such an application, vertices and/or edges of the graph may not be perfectly reliable; an agent may experience failure or a communication link rendered inoperable. With the goal of designing robust swarm formations, or unit ball graphs with high reliability (probability of connectedness), in a preliminary conference paper we provided an algorithm with cubic time complexity to determine all possible changes to a unit ball graph by repositioning a single vertex. Using this algorithm and Monte Carlo simulations, one obtains an efficient method to modify a unit ball graph by moving a single vertex to a location which maximizes the reliability. Another important consideration in many swarm missions is area coverage, yet highly reliable ball graphs often contain clusters of vertices. Here, we generalize our previous algorithm to improve area coverage as well as reliability. Our algorithm determines a location to add or move a vertex within a unit ball graph which maximizes the reliability, under the constraint that no other vertices of the graph be within some fixed distance. We compare this method of obtaining graphs with high reliability and evenly distributed area coverage to another method which uses a modified Fruchterman-Reingold algorithm for ball graphs.

cs.DC↗

arXiv2025

Multi-Exit Kolmogorov-Arnold Networks: enhancing accuracy and parsimony

Kolmogorov-Arnold Networks (KANs) uniquely combine high accuracy with interpretability, making them valuable for scientific modeling. However, it is unclear a priori how deep a network needs to be for any given task, and deeper KANs can be difficult to optimize and interpret. Here we introduce multi-exit KANs, where each layer includes its own prediction branch, enabling the network to make accurate predictions at multiple depths simultaneously. This architecture provides deep supervision that improves training while discovering the right level of model complexity for each task. Multi-exit KANs consistently outperform standard, single-exit versions on synthetic functions, dynamical systems, and real-world datasets. Remarkably, the best predictions often come from earlier, simpler exits, revealing that these networks naturally identify smaller, more parsimonious and interpretable models without sacrificing accuracy. To automate this discovery, we develop a differentiable "learning-to-exit" algorithm that balances contributions from exits during training. Our approach offers scientists a practical way to achieve both high performance and interpretability, addressing a fundamental challenge in machine learning for scientific discovery.

cs.LG↗

arXiv2025

[RETRACTED]Evolving Form and Function: Dual-Objective Optimization in Neural Symbolic Regression Networks

[RETRACTED]Data increasingly abounds, but distilling their underlying relationships down to something interpretable remains challenging. One approach is genetic programming, which `symbolically regresses' a data set down into an equation. However, symbolic regression (SR) faces the issue of requiring training from scratch for each new dataset. To generalize across all datasets, deep learning techniques have been applied to SR. These networks, however, are only able to be trained using a symbolic objective: NN-generated and target equations are symbolically compared. But this does not consider the predictive power of these equations, which could be measured by a behavioral objective that compares the generated equation's predictions to actual data. Here we introduce a method that combines gradient descent and evolutionary computation to yield neural networks that minimize the symbolic and behavioral errors of the equations they generate from data. As a result, these evolved networks are shown to generate more symbolically and behaviorally accurate equations than those generated by networks trained by state-of-the-art gradient based neural symbolic regression methods. We hope this method suggests that evolutionary algorithms, combined with gradient descent, can improve SR results by yielding equations with more accurate form and function.

cs.NE↗

arXiv2024

Data-driven Modeling of Granular Chains with Modern Koopman Theory

Externally driven dense packings of particles can exhibit nonlinear wave phenomena that are not described by effective medium theory or linearized approximate models. Such nontrivial wave responses can be exploited to design sound-focusing/scrambling devices, acoustic filters, and analog computational units. At high amplitude vibrations or low confinement pressures, the effect of nonlinear particle contacts becomes increasingly noticeable, and the interplay of nonlinearity, disorder, and discreteness in the system gives rise to remarkable properties, particularly useful in designing structures with exotic properties. In this paper, we build upon the data-driven methods in dynamical system analysis and show that the Koopman spectral theory can be applied to granular crystals, enabling their phase space analysis beyond the linearizable regime and without recourse to any approximations considered in the previous works. We show that a deep neural network can map the dynamics to a latent space where the essential nonlinearity of the granular system unfolds into a high-dimensional linear space. As a proof of concept, we use data from numerical simulations of a two-particle system and evaluate the accuracy of the trajectory predictions under various initial conditions. By incorporating data from experimental measurements, our proposed framework can directly capture the underlying dynamics without imposing any assumptions about the physics model. Spectral analysis of the trained surrogate system can help bridge the gap between the simulation results and the physical realization of granular crystals and facilitate the inverse design of materials with desired behaviors.

cond-mat.soft↗

arXiv2023

Revisiting Cont's Stylized Facts for Modern Stock Markets

In 2001, Rama Cont introduced a now-widely used set of 'stylized facts' to synthesize empirical studies of financial price changes (returns), resulting in 11 statistical properties common to a large set of assets and markets. These properties are viewed as constraints a model should be able to reproduce in order to accurately represent returns in a market. It has not been established whether the characteristics Cont noted in 2001 still hold for modern markets following significant regulatory shifts and technological advances. It is also not clear whether a given time series of financial returns for an asset will express all 11 stylized facts. We test both of these propositions by attempting to replicate each of Cont's 11 stylized facts for intraday returns of the individual stocks in the Dow 30, using the same authoritative data as that used by the U.S. regulator from October 2018 - March 2019. We find conclusive evidence for eight of Cont's original facts and no support for the remaining three. Our study represents the first test of Cont's 11 stylized facts against a consistent set of stocks, therefore providing insight into how these stylized facts should be viewed in the context of modern stock markets.

q-fin.ST↗

arXiv2023

Measuring Centralization of Online Platforms Through Size and Interconnection of Communities

Decentralized architecture offers a robust and flexible structure for online platforms, since centralized moderation and computation can be easy to disrupt with targeted attacks. However, a platform offering a decentralized architecture does not guarantee that users will use it in a decentralized way, and measuring the centralization of socio-technical networks is not an easy task. In this paper we introduce a method of characterizing community influence in terms of how many edges between communities would be disrupted by a community's removal. Our approach provides a careful definition of "centralization" appropriate in bipartite user-community socio-technical networks, and demonstrates the inadequacy of more trivial methods for interrogating centralization such as examining the distribution of community sizes. We use this method to compare the structure of multiple socio-technical platforms -- Mastodon, git code hosting servers, BitChute, Usenet, and Voat -- and find a range of structures, from interconnected but decentralized git servers to an effectively centralized use of Mastodon servers, as well as multiscale hybrid network structures of disconnected Voat subverses. As the ecosystem of socio-technical platforms diversifies, it becomes critical to not solely focus on the underlying technologies but also consider the structure of how users interact through the technical infrastructure.

cs.SI↗

arXiv2022

Network Cards: concise, readable summaries of network data

The deluge of network datasets demands a standard way to effectively and succinctly summarize network datasets. Building on similar efforts to standardize the documentation of models and datasets in machine learning, here we propose network cards, short summaries of network datasets that can capture not only the basic statistics of the network but also information about the data construction process, provenance, ethical considerations, and other metadata. In this paper, we lay out (1) the rationales and objectives for network cards, (2) key elements that should be included in network cards, and (3) example network cards to underscore their benefits across a variety of research domains. We also provide a schema, templates, and a software package for generating network cards.

cs.SI↗

arXiv2022

The OCEAN mailing list data set: Network analysis spanning mailing lists and code repositories

Communication surrounding the development of an open source project largely occurs outside the software repository itself. Historically, large communities often used a collection of mailing lists to discuss the different aspects of their projects. Multimodal tool use, with software development and communication happening on different channels, complicates the study of open source projects as a sociotechnical system. Here, we combine and standardize mailing lists of the Python community, resulting in 954,287 messages from 1995 to the present. We share all scraping and cleaning code to facilitate reproduction of this work, as well as smaller datasets for the Golang (122,721 messages), Angular (20,041 messages) and Node.js (12,514 messages) communities. To showcase the usefulness of these data, we focus on the CPython repository and merge the technical layer (which GitHub account works on what file and with whom) with the social layer (messages from unique email addresses) by identifying 33% of GitHub contributors in the mailing list data. We then explore correlations between the valence of social messaging and the structure of the collaboration network. We discuss how these data provide a laboratory to test theories from standard organizational science in large open source projects.

cs.CY↗

arXiv2022

Hierarchical team structure and multidimensional localization (or siloing) on networks

Knowledge silos emerge when structural properties of organizational interaction networks limit the diffusion of information. These structural barriers are known to take many forms at different scales - hubs in otherwise sparse organisations, large dense teams, or global core-periphery structure - but we lack an understanding of how these different structures interact. Here we bridge the gap between the mathematical literature on localization of spreading dynamics and the more applied literature on knowledge silos in organizational interaction networks. To do so, we introduce a new model that considers a layered structure of teams to unveil a new form of hierarchical localization (i.e., the localization of information at the top or center of an organization) and study its interplay with known phenomena of mesoscopic localization (i.e., the localization of information in large groups), $k$-core localization (i.e., around denser $k$-cores) and hub localization (i.e., around high degree stars). We also include a complex contagion mechanism by considering a general infection kernel which can depend on hierarchical level (influence), degree (popularity), infectious neighbors (social reinforcement) or team size (importance). This general model allows us to study the multifaceted phenomenon of information siloing in complex organizational interaction networks and opens the door to new optimization problems to promote or hinder the emergence of different localization regimes.

physics.soc-ph↗

arXiv2021

The penumbra of open source: projects outside of centralized platforms are longer maintained, more academic and more collaborative

GitHub has become the central online platform for much of open source, hosting most open source code repositories. With this popularity, the public digital traces of GitHub are now a valuable means to study teamwork and collaboration. In many ways, however, GitHub is a convenience sample, and may not be representative of open source development off the platform. Here we develop a novel, extensive sample of public open source project repositories outside of centralized platforms. We characterized these projects along a number of dimensions, and compare to a time-matched sample of corresponding GitHub projects. Our sample projects tend to have more collaborators, are maintained for longer periods, and tend to be more focused on academic and scientific problems.

cs.CY↗

arXiv2019

Accurate inference of crowdsourcing properties when using efficient allocation strategies

Allocation strategies improve the efficiency of crowdsourcing by decreasing the work needed to complete individual tasks accurately. However, these algorithms introduce bias by preferentially allocating workers onto easy tasks, leading to sets of completed tasks that are no longer representative of all tasks. This bias challenges inference of problem-wide properties such as typical task difficulty or crowd properties such as worker completion times, important information that goes beyond the crowd responses themselves. Here we study inference about problem properties when using an allocation algorithm to improve crowd efficiency. We introduce Decision-Explicit Probability Sampling (DEPS), a novel method to perform inference of problem properties while accounting for the potential bias introduced by an allocation strategy. Experiments on real and synthetic crowdsourcing data show that DEPS outperforms baseline inference methods while still leveraging the efficiency gains of the allocation method. The ability to perform accurate inference of general properties when using non-representative data allows crowdsourcers to extract more knowledge out of a given crowdsourced dataset.

cs.LG↗

arXiv2006

Network Structure Revealed by Short Cycles

This article explores the relationship between communities and short cycles in complex networks, based on the fact that nodes more densely connected amongst one another are more likely to be linked through short cycles. By identifying combinations of 3-, 4- and 5-edge-cycles, a subnetwork is obtained which contains only those nodes and links belonging to such cycles, which can then be used to highlight community structure. Examples are shown using a theoretical model (Sznajd networks) and a real-world network (NCAA football).

cond-mat.dis-nn↗