arXiv subjects

Ana Trisovic

Publications and source records attributed to Ana Trisovic.

13 recordsLinked to original sources

arXiv2026

WFDroneBench: A Benchmark for Sensor Placement and Drone Routing for Wildfire Detection

Increasingly frequent and severe wildfires threaten ecosystems, public health, and infrastructure. Early detection is vital but limited by existing monitoring systems. Drones offer mobile, real-time coverage, but optimizing sensor placement and drone routing in dynamic fire zones remains challenging. To address this, we introduce WFDroneBench, an open-source Python benchmarking library for early wildfire detection that integrates machine-learned risk maps with optimization-based deployment strategies for sensors, charging stations, and drones. It evaluates risk maps, optimization strategies, and monitoring equipment using standardized metrics and realistic wildfire simulations. The framework supports benchmarking across predictive and decision-making components: machine learning researchers can assess risk models and compare routing strategies. WFDroneBench includes 7746 scenarios across 49 locations, built from historical ignitions, real-world wildfire risk maps, and simulated fire spread, along with two ground detector and three drone routing strategies. Our experiments show that the risk-aware strategy Max-Coverage significantly outperforms other baselines when risk maps are sufficiently accurate, achieving the fastest detection on the most difficult fires. We further find that risk-aware static infrastructure helps even under an imperfect risk map and drone-based detection outperforms ground sensors. Finally, our results reveal two key open challenges: (i) detecting small fires rapidly and reliably, and (ii) improving risk-map prediction, where the gap between ground-truth ignition patterns and available risk maps highlights a significant opportunity for ML innovation. We openly release all code, data, and documentation.

math.OC↗

arXiv2026

A Causal DAG Prior for Synthetic Time-Series Classification Datasets

A Prior-data fitted Network learns the posterior predictive induced by its training prior; bringing this paradigm to multivariate time-series classification therefore calls for a synthetic generator that produces complete labelled datasets with temporal structure. We introduce a causal prior that synthesizes each dataset from a randomly sampled DAG over typed nodes across two modalities (tabular attributes and time series), natively producing multivariate, multi-class TSC datasets with cross-modal causal structure across channels, timesteps and labels, a regime not addressed by existing synthetic priors. To validate the prior, we finetune TabPFN v2.5 with minimal adaptations and evaluate on 75 UCR/UEA datasets within TabPFN's operating regime. Finetuning on our generator significantly outperforms both the unmodified upstream model and a tabular-only ablation of the same prior (Wilcoxon signed-rank $p=3.0\times 10^{-8}$ on ROC-AUC), isolating the contribution of the cross-modal temporal structure.

cs.LG↗

arXiv2026

RocketPFN: Accurate Time Series Classification via In-Context Learning

We introduce RocketPFN, a training-free pipeline for time series classification that combines random convolutional feature extraction (Rocket) with in-context classification via a pretrained tabular foundation model (TabPFN v2.5). On 92 UCR datasets (30-resample protocol), RocketPFN matches HC2, the strongest published method on the archive, in mean accuracy (both 0.900, Wilcoxon p=0.50), with no training on the target data and a median inference time of 30 seconds per fold. It also significantly outperforms every individual classifier in the HC2 ensemble. On UEA (20 datasets) the difference is likewise not statistically significant. A separate comparison concerns TSC foundation models: when paired with the same downstream classifier, MOMENT, Mantis, and MantisV2 are all significantly outperformed by RocketPFN using fewer extracted features and no learned parameters (p<0.001 in each case). This holds even when the encoders were pretrained on corpora that include the UCR training samples. We propose this two-stage pipeline as a reference point for evaluating zero-shot TSC foundation models.

cs.LG↗

arXiv2025

On the Origin of Algorithmic Progress in AI

Algorithms have been estimated to increase AI training FLOP efficiency by a factor of 22,000 between 2012 and 2023 [Ho et al., 2024]. Running small-scale ablation experiments on key innovations from this time period, we are able to account for less than 10x of these gains. Surveying the broader literature, we estimate that additional innovations not included in our ablations account for less than 10x, yielding a total under 100x. This leads us to conduct scaling experiments, which reveal that much of this efficiency gap can be explained by algorithms with scale-dependent efficiency improvements. In particular, we conduct scaling experiments between LSTMs and Transformers, finding exponent differences in their compute-optimal scaling law while finding little scaling difference for many other innovations. These experiments demonstrate that - contrary to standard assumptions - an algorithm's efficiency gains are tied to compute scale. Using experimental extrapolation and literature estimates, we account for 6,930x efficiency gains over the same time period, with the scale-dependent LSTM-to-Transformer transition accounting for the majority of gains. Our results indicate that algorithmic progress for small models has been far slower than previously assumed, and that measures of algorithmic efficiency are strongly reference-dependent.

cs.LG↗

arXiv2023

SpaCE: The Spatial Confounding Environment

Spatial confounding poses a significant challenge in scientific studies involving spatial data, where unobserved spatial variables can influence both treatment and outcome, possibly leading to spurious associations. To address this problem, we introduce SpaCE: The Spatial Confounding Environment, the first toolkit to provide realistic benchmark datasets and tools for systematically evaluating causal inference methods designed to alleviate spatial confounding. Each dataset includes training data, true counterfactuals, a spatial graph with coordinates, and smoothness and confounding scores characterizing the effect of a missing spatial confounder. It also includes realistic semi-synthetic outcomes and counterfactuals, generated using state-of-the-art machine learning ensembles, following best practices for causal inference benchmarks. The datasets cover real treatment and covariates from diverse domains, including climate, health and social sciences. SpaCE facilitates an automated end-to-end pipeline, simplifying data loading, experimental setup, and evaluating machine learning and causal inference models. The SpaCE project provides several dozens of datasets of diverse sizes and spatial complexity. It is publicly available as a Python package, encouraging community feedback and contributions.

cs.LG↗

arXiv2023

Cluster Analysis of Open Research Data and a Case for Replication Metadata

Research data are often released upon journal publication to enable result verification and reproducibility. For that reason, research dissemination infrastructures typically support diverse datasets coming from numerous disciplines, from tabular data and program code to audio-visual files. Metadata, or data about data, is critical to making research outputs adequately documented and FAIR. Aiming to contribute to the discussions on the development of metadata for research outputs, I conducted an exploratory analysis to determine how research datasets cluster based on what researchers organically deposit together. I use the content of over 40,000 datasets from the Harvard Dataverse research data repository as my sample for the cluster analysis. I find that the majority of the clusters are formed by single-type datasets, while in the rest of the sample, no meaningful clusters can be identified. For the result interpretation, I use the metadata standard employed by DataCite, a leading organization for documenting a scholarly record, and map existing resource types to my results. About 65% of the sample can be described with a single-type metadata (such as Dataset, Software or Report), while the rest would require aggregate metadata types. Though DataCite supports an aggregate type such as a Collection, I argue that a significant number of datasets, in particular those containing both data and code files (about 20% of the sample) would be more accurately described as a Replication resource metadata type. Such resource type would be particularly useful in facilitating research reproducibility.

cs.DL↗

arXiv2023

Advancing Software Citation Implementation (Software Citation Workshop 2022)

Software is foundationally important to scientific and social progress, however, traditional acknowledgment of the use of others' work has not adapted in step with the rapid development and use of software in research. This report outlines a series of collaborative discussions that brought together an international group of stakeholders and experts representing many communities, forms of labor, and expertise. Participants addressed specific challenges about software citation that have so far gone unresolved. The discussions took place in summer 2022 both online and in-person and involved a total of 51 participants. The activities described in this paper were intended to identify and prioritize specific software citation problems, develop (potential) interventions, and lay out a series of mutually supporting approaches to address them. The outcomes of this report will be useful for the GLAM (Galleries, Libraries, Archives, Museums) community, repository managers and curators, research software developers, and publishers.

cs.DL↗

arXiv2021

Toward Reusable Science with Readable Code and Reproducibility

An essential part of research and scientific communication is researchers' ability to reproduce the results of others. While there have been increasing standards for authors to make data and code available, many of these files are hard to re-execute in practice, leading to a lack of research reproducibility. This poses a major problem for students and researchers in the same field who cannot leverage the previously published findings for study or further inquiry. To address this, we propose an open-source platform named RE3 that helps improve the reproducibility and readability of research projects involving R code. Our platform incorporates assessing code readability with a machine learning model trained on a code readability survey and an automatic containerization service that executes code files and warns users of reproducibility errors. This process helps ensure the reproducibility and readability of projects and therefore fast-track their verification and reuse.

cs.DL↗

arXiv2021

Towards FAIR Principles for Open Hardware

The lack of scientific openness is identified as one of the key challenges of computational reproducibility. In addition to Open Data, Free and Open-source Software (FOSS) and Open Hardware (OH) can address this challenge by introducing open policies, standards, and recommendations. However, while both FOSS and OH are free to use, study, modify, and redistribute, there are significant differences in sharing and reusing these artifacts. FOSS is increasingly supported with software repositories, but support for OH is lacking, potentially due to the complexity of its digital format and licensing. This paper proposes leveraging FAIR principles to make OH findable, accessible, interoperable, and reusable. We define what FAIR means for OH, how it differs from FOSS, and present examples of unique demands. Also, we evaluate dissemination platforms currently used for OH and provide recommendations.

cs.DL↗

arXiv2021

Packaging research artefacts with RO-Crate

An increasing number of researchers support reproducibility by including pointers to and descriptions of datasets, software and methods in their publications. However, scientific articles may be ambiguous, incomplete and difficult to process by automated systems. In this paper we introduce RO-Crate, an open, community-driven, and lightweight approach to packaging research artefacts along with their metadata in a machine readable manner. RO-Crate is based on Schema$.$org annotations in JSON-LD, aiming to establish best practices to formally describe metadata in an accessible and practical way for their use in a wide variety of situations. An RO-Crate is a structured archive of all the items that contributed to a research outcome, including their identifiers, provenance, relations and annotations. As a general purpose packaging approach for data and their metadata, RO-Crate is used across multiple areas, including bioinformatics, digital humanities and regulatory sciences. By applying "just enough" Linked Data standards, RO-Crate simplifies the process of making research outputs FAIR while also enhancing research reproducibility. An RO-Crate for this article is available at https://w3id.org/ro/doi/10.5281/zenodo.5146227

cs.DL↗

arXiv2021

A large-scale study on research code quality and execution

This article presents a study on the quality and execution of research code from publicly-available replication datasets at the Harvard Dataverse repository. Research code is typically created by a group of scientists and published together with academic papers to facilitate research transparency and reproducibility. For this study, we define ten questions to address aspects impacting research reproducibility and reuse. First, we retrieve and analyze more than 2000 replication datasets with over 9000 unique R files published from 2010 to 2020. Second, we execute the code in a clean runtime environment to assess its ease of reuse. Common coding errors were identified, and some of them were solved with automatic code cleaning to aid code execution. We find that 74\% of R files crashed in the initial execution, while 56\% crashed when code cleaning was applied, showing that many errors can be prevented with good coding practices. We also analyze the replication datasets from journals' collections and discuss the impact of the journal policy strictness on the code re-execution rate. Finally, based on our results, we propose a set of recommendations for code dissemination aimed at researchers, journals, and repositories.

cs.SE↗

arXiv2020

Advancing computational reproducibility in the Dataverse data repository platform

Recent reproducibility case studies have raised concerns showing that much of the deposited research has not been reproducible. One of their conclusions was that the way data repositories store research data and code cannot fully facilitate reproducibility due to the absence of a runtime environment needed for the code execution. New specialized reproducibility tools provide cloud-based computational environments for code encapsulation, thus enabling research portability and reproducibility. However, they do not often enable research discoverability, standardized data citation, or long-term archival like data repositories do. This paper addresses the shortcomings of data repositories and reproducibility tools and how they could be overcome to improve the current lack of computational reproducibility in published and archived research outputs.

cs.DL↗

arXiv2020

Provenance tracking in the LHCb software

Even though computational reproducibility is widely accepted as necessary for research validation and reuse, it is often not considered during the research process. This is because reproducibility tools are typically stand-alone and require additional training to be employed. In this article, we present a solution to foster reproducibility, which is integrated within existing scientific software that is actively used in the LHCb collaboration. Our provenance tracking service captures metadata of a dataset, which is then saved inside the output data file on the disk. The captured information allows a complete understanding of how the file was produced and enables a user to reproduce the dataset, even when the original input code (that was used to initially produce the dataset) is altered or lost. This article describes the implementation of the service and gives examples of its application.

cs.DC↗