arXiv subjects

Roberto Gonzalez

Publications and source records attributed to Roberto Gonzalez.

14 recordsLinked to original sources

arXiv2026

vAccSOL: Efficient and Transparent AI Vision Offloading for Mobile Robots

Mobile robots are increasingly deployed for inspection, patrol, and search-and-rescue operations, relying on computer vision for perception, navigation, and autonomous decision-making. However, executing modern vision workloads onboard is challenging due to limited compute resources and strict energy constraints. While some platforms include embedded accelerators, these are typically tied to proprietary software stacks, leaving user-defined workloads to run on resource-constrained companion computers. We present vAccSOL, a framework for efficient and transparent execution of AI-based vision workloads across heterogeneous robotic and edge platforms. vAccSOL integrates two components: SOL, a neural network compiler that generates optimized inference libraries with minimal runtime dependencies, and vAccel, a lightweight execution framework that transparently dispatches inference locally on the robot or to nearby edge infrastructure. This combination enables hardware-optimized inference and flexible execution placement without requiring modifications to robot applications. We evaluate vAccSOL on a real-world testbed with a commercial quadruped robot and twelve deep learning models covering image classification, video classification, and semantic segmentation. Compared to a PyTorch compiler baseline, SOL achieves comparable or better inference performance. With edge offloading, vAccSOL reduces robot-side power consumption by up to 80% and edge-side power by up to 60% compared to PyTorch, while increasing vision pipeline frame rate by up to 24x, extending the operating lifetime of battery-powered robots.

cs.RO↗

arXiv2024

Clean Up the Mess: Addressing Data Pollution in Cryptocurrency Abuse Reporting Services

Cryptocurrency abuse reporting services are a valuable data source about abusive blockchain addresses, prevalent types of cryptocurrency abuse, and their financial impact on victims. However, they may suffer data pollution due to their crowd-sourced nature. This work analyzes the extent and impact of data pollution in cryptocurrency abuse reporting services and proposes a novel LLM-based defense to address the pollution. We collect 289K abuse reports submitted over 6 years to two popular services and use them to answer three research questions. RQ1 analyzes the extent and impact of pollution. We show that spam reports will eventually flood unchecked abuse reporting services, with BitcoinAbuse receiving 75% of spam before stopping operations. We build a public dataset of 19,443 abuse reports labeled with 19 popular abuse types and use it to reveal the inaccuracy of user-reported abuse types. We identified 91 (0.1%) benign addresses reported, responsible for 60% of all the received funds. RQ2 examines whether we can automate identifying valid reports and their classification into abuse types. We propose an unsupervised LLM-based classifier that achieves an F1 score of 0.95 when classifying reports, an F1 of 0.89 when classifying out-of-distribution data, and an F1 of 0.99 when identifying spam reports. Our unsupervised LLM-based classifier clearly outperforms two baselines: a supervised classifier and a naive usage of the LLM. Finally, RQ3 demonstrates the usefulness of our LLM-based classifier for quantifying the financial impact of different cryptocurrency abuse types. We show that victim-reported losses heavily underestimate cybercriminal revenue by estimating a 29 times higher revenue from deposit transactions. We identified that investment scams have the highest financial impact and that extortions have lower conversion rates but compensate for them with massive email campaigns.

cs.CR↗

arXiv2023

Time for aCTIon: Automated Analysis of Cyber Threat Intelligence in the Wild

Cyber Threat Intelligence (CTI) plays a crucial role in assessing risks and enhancing security for organizations. However, the process of extracting relevant information from unstructured text sources can be expensive and time-consuming. Our empirical experience shows that existing tools for automated structured CTI extraction have performance limitations. Furthermore, the community lacks a common benchmark to quantitatively assess their performance. We fill these gaps providing a new large open benchmark dataset and aCTIon, a structured CTI information extraction tool. The dataset includes 204 real-world publicly available reports and their corresponding structured CTI information in STIX format. Our team curated the dataset involving three independent groups of CTI analysts working over the course of several months. To the best of our knowledge, this dataset is two orders of magnitude larger than previously released open source datasets. We then design aCTIon, leveraging recently introduced large language models (GPT3.5) in the context of two custom information extraction pipelines. We compare our method with 10 solutions presented in previous work, for which we develop our own implementations when open-source implementations were lacking. Our results show that aCTIon outperforms previous work for structured CTI extraction with an improvement of the F1-score from 10%points to 50%points across all tasks.

cs.CR↗

arXiv2023

Redshift evolution of the dark matter haloes shapes

In this work, we aim at investigating the morphology evolution of Milky Way mass-like dark matter haloes selected from the CIELO and IllustrisTNG Projects. The connection between halo shapes and their environment has been studied in previous works at z=0 but their connection remains yet to be fully understood. We focus on the evolution across cosmic time of the halo shapes and the relation with the infalling material, using hydrodynamical simulations. Our findings show that haloes tend to be more triaxial at earlier times as a consequence of stronger accretion in the direction of the filaments. As the haloes evolve towards a dominant isotropic accretion mode and relaxation, their shape at 20 percent of the virial mass becomes more spherical. In agreement with previous results, baryons have an important effect within the inner regions of the haloes, driving them from triaxial to rounder shapes. We also find a correlation between the strength of the quadrupole infalling mode and the degree of ellipticity of the haloes: as the filament strength decreases steadily with redshift, the haloes became more spherical and less elliptical.

astro-ph.GA↗

arXiv2022

syslrn: Learning What to Monitor for Efficient Anomaly Detection

While monitoring system behavior to detect anomalies and failures is important, existing methods based on log-analysis can only be as good as the information contained in the logs, and other approaches that look at the OS-level software state introduce high overheads. We tackle the problem with syslrn, a system that first builds an understanding of a target system offline, and then tailors the online monitoring instrumentation based on the learned identifiers of normal behavior. While our syslrn prototype is still preliminary and lacks many features, we show in a case study for the monitoring of OpenStack failures that it can outperform state-of-the-art log-analysis systems with little overhead.

cs.LG↗

arXiv2018

TransRev: Modeling Reviews as Translations from Users to Items

The text of a review expresses the sentiment a customer has towards a particular product. This is exploited in sentiment analysis where machine learning models are used to predict the review score from the text of the review. Furthermore, the products costumers have purchased in the past are indicative of the products they will purchase in the future. This is what recommender systems exploit by learning models from purchase information to predict the items a customer might be interested in. We propose TransRev, an approach to the product recommendation problem that integrates ideas from recommender systems, sentiment analysis, and multi-relational learning into a joint learning objective. TransRev learns vector representations for users, items, and reviews. The embedding of a review is learned such that (a) it performs well as input feature of a regression model for sentiment prediction; and (b) it always translates the reviewer embedding to the embedding of the reviewed items. This allows TransRev to approximate a review embedding at test time as the difference of the embedding of each item and the user embedding. The approximated review embedding is then used with the regression model to predict the review score for each item. TransRev outperforms state of the art recommender systems on a large number of benchmark data sets. Moreover, it is able to retrieve, for each user and item, the review text from the training set whose embedding is most similar to the approximated review embedding.

cs.IR↗

arXiv2017

Net2Vec: Deep Learning for the Network

We present Net2Vec, a flexible high-performance platform that allows the execution of deep learning algorithms in the communication network. Net2Vec is able to capture data from the network at more than 60Gbps, transform it into meaningful tuples and apply predictions over the tuples in real time. This platform can be used for different purposes ranging from traffic classification to network performance analysis. Finally, we showcase the use of Net2Vec by implementing and testing a solution able to profile network users at line rate using traces coming from a real network. We show that the use of deep learning for this case outperforms the baseline method both in terms of accuracy and performance.

cs.NI↗

arXiv2017

Tracing the cosmic web

The cosmic web is one of the most striking features of the distribution of galaxies and dark matter on the largest scales in the Universe. It is composed of dense regions packed full of galaxies, long filamentary bridges, flattened sheets and vast low density voids. The study of the cosmic web has focused primarily on the identification of such features, and on understanding the environmental effects on galaxy formation and halo assembly. As such, a variety of different methods have been devised to classify the cosmic web -- depending on the data at hand, be it numerical simulations, large sky surveys or other. In this paper we bring twelve of these methods together and apply them to the same data set in order to understand how they compare. In general these cosmic web classifiers have been designed with different cosmological goals in mind, and to study different questions. Therefore one would not {\it a priori} expect agreement between different techniques however, many of these methods do converge on the identification of specific features. In this paper we study the agreements and disparities of the different methods. For example, each method finds that knots inhabit higher density regions than filaments, etc. and that voids have the lowest densities. For a given web environment, we find substantial overlap in the density range assigned by each web classification scheme. We also compare classifications on a halo-by-halo basis; for example, we find that 9 of 12 methods classify around a third of group-mass haloes (i.e. $M_{\rm halo}\sim10^{13.5}h^{-1}M_{\odot}$) as being in filaments. Lastly, so that any future cosmic web classification scheme can be compared to the 12 methods used here, we have made all the data used in this paper public.

astro-ph.CO↗

arXiv2017

Understanding the evolution of multimedia content in the Internet through BitTorrent glasses

Today's Internet traffic is mostly dominated by multimedia content and the prediction is that this trend will intensify in the future. Therefore, main Internet players, such as ISPs, content delivery platforms (e.g. Youtube, Bitorrent, Netflix, etc) or CDN operators, need to understand the evolution of multimedia content availability and popularity in order to adapt their infrastructures and resources to satisfy clients requirements while they minimize their costs. This paper presents a thorough analysis on the evolution of multimedia content available in BitTorrent. Specifically, we analyze the evolution of four relevant metrics across different content categories: content availability, content popularity, content size and user's feedback. To this end we leverage a large-scale dataset formed by 4 snapshots collected from the most popular BitTorrent portal, namely The Pirate Bay, between Nov. 2009 and Feb. 2012. Overall our dataset is formed by more than 160k content that attracted more than 185M of download sessions.

cs.NI↗

arXiv2016

Understanding the Detection of View Fraud in Video Content Portals

While substantial effort has been devoted to understand fraudulent activity in traditional online advertising (search and banner), more recent forms such as video ads have received little attention. The understanding and identification of fraudulent activity (i.e., fake views) in video ads for advertisers, is complicated as they rely exclusively on the detection mechanisms deployed by video hosting portals. In this context, the development of independent tools able to monitor and audit the fidelity of these systems are missing today and needed by both industry and regulators. In this paper we present a first set of tools to serve this purpose. Using our tools, we evaluate the performance of the audit systems of five major online video portals. Our results reveal that YouTube's detection system significantly outperforms all the others. Despite this, a systematic evaluation indicates that it may still be susceptible to simple attacks. Furthermore, we find that YouTube penalizes its videos' public and monetized view counters differently, the former being more aggressive. This means that views identified as fake and discounted from the public view counter are still monetized. We speculate that even though YouTube's policy puts in lots of effort to compensate users after an attack is discovered, this practice places the burden of the risk on the advertisers, who pay to get their ads displayed.

cs.CY↗

arXiv2013

Google+ or Google-?: Dissecting the Evolution of the New OSN in its First Year

In the era when Facebook and Twitter dominate the market for social media, Google has introduced Google+ (G+) and reported a significant growth in its size while others called it a ghost town. This begs the question that "whether G+ can really attract a significant number of connected and active users despite the dominance of Facebook and Twitter?". This paper tackles the above question by presenting a detailed characterization of G+ based on large scale measurements. We identify the main components of G+ structure, characterize the key features of their users and their evolution over time. We then conduct detailed analysis on the evolution of connectivity and activity among users in the largest connected component (LCC) of G+ structure, and compare their characteristics with other major OSNs. We show that despite the dramatic growth in the size of G+, the relative size of LCC has been decreasing and its connectivity has become less clustered. While the aggregate user activity has gradually increased, only a very small fraction of users exhibit any type of activity. To our knowledge, our study offers the most comprehensive characterization of G+ based on the largest collected data sets.

cs.SI↗

arXiv2012

TorrentGuard: stopping scam and malware distribution in the BitTorrent ecosystem

In this paper we conduct a large scale measurement study in order to analyse the fake content publishing phenomenon in the BitTorrent Ecosystem. Our results reveal that fake content represents an important portion (35%) of those files shared in BitTorrent and just a few tens of users are responsible for 90% of this content. Furthermore, more than 99% of the analysed fake files are linked to either malware or scam websites. This creates a serious threat for the BitTorrent ecosystem. To address this issue, we present a new detection tool named TorrentGuard for the early detection of fake content. Based on our evaluation this tool may prevent the download of more than 35 millions of fake files per year. This could help to reduce the number of computer infections and scams suffered by BitTorrent users. TorrentGuard is already available and it can be accessed through both a webpage or a Vuze plugin.

cs.CR↗

arXiv2011

Where are my followers? Understanding the Locality Effect in Twitter

Twitter is one of the most used applications in the current Internet with more than 200M accounts created so far. As other large-scale systems Twitter can obtain enefit by exploiting the Locality effect existing among its users. In this paper we perform the first comprehensive study of the Locality effect of Twitter. For this purpose we have collected the geographical location of around 1M Twitter users and 16M of their followers. Our results demonstrate that language and cultural characteristics determine the level of Locality expected for different countries. Those countries with a different language than English such as Brazil typically show a high intra-country Locality whereas those others where English is official or co-official language suffer from an external Locality effect. This is, their users have a larger number of followers in US than within their same country. This is produced by two reasons: first, US is the dominant country in Twitter counting with around half of the users, and second, these countries share a common language and cultural characteristics with US.

cs.SI↗

arXiv2010

Local and global environmental effects on galaxies and active galactic nuclei

We study the properties of SDSS galaxies with and without AGN detection as a function of the local and global environment measured via the local density, the mass of the galaxy host group (parameterised by the group luminosity) and distance to massive clusters. Our results can be divided in two main subjects, the environments of galaxies and their relation to the assembly of their host haloes, and the environments of AGN. (i) For the full SDSS sample, we find indications that the local galaxy density is the most efficient parameter to separate galaxy populations, but we also find that galaxies at fixed local density show some remaining variation of their properties as a function of the distance to the nearest cluster of galaxies (in a range of 0 to 10 cluster virial radii). These differences seem to become less significant if the galaxy samples are additionally constrained to be hosted by groups of similar total luminosity. (ii) In AGN host galaxies, the morphology-density relation is much less noticeable when compared to the behaviour of the full SDSS sample. In order to interpret this result we analyse control samples constructed using galaxies with no detected AGN activity with matching distributions of redshifts, stellar masses, r-band luminosities, g-r colours, concentrations, local densities, host group luminosities, and fractions of central and satellite galaxies. The control samples also show a similar small dependence on the local density indicating an influence from the AGN selection, but their colours are slightly bluer compared to the AGN hosts regardless of local density. Furthermore, even when the local density is held fixed at intermediate or high values, and the distance to the closest cluster of galaxies is allowed to vary, AGN control galaxies away from clusters tend to be bluer than the AGN hosts. (ABRIDGED)

astro-ph.CO↗