Search

Tabular data is ubiquitous in many real-life systems. In particular, time-dependent tabular data, where rows are chronologically related, is typically used for recording historical events, e.g., financial transactions, healthcare records, or stock history. Recently, hierarchical variants of the attention mechanism of transformer architectures have been used to model tabular time-series data. At first, rows (or columns) are encoded separately by computing attention between their fields. Subsequently, encoded rows (or columns) are attended to one another to model the entire tabular time-series. While efficient, this approach constrains the attention granularity and limits its ability to learn patterns at the field-level across separate rows, or columns. We take a first step to address this gap by proposing Fieldy, a fine-grained hierarchical model that contextualizes fields at both the row and column levels. We compare our proposal against state of the art models on regression and classification tasks using public tabular time-series datasets. Our results show that combining row-wise and column-wise attention improves performance without increasing model size. Code and data are available at https://github.com/raphaaal/fieldy., Comment: 9 pages; Camera Ready version

Diffusion models for Text-to-Image (T2I) conditional generation have seen tremendous success recently. Despite their success, accurately capturing user intentions with these models still requires a laborious trial and error process. This challenge is commonly identified as a model alignment problem, an issue that has attracted considerable attention by the research community. Instead of relying on fine-grained linguistic analyses of prompts, human annotation, or auxiliary vision-language models to steer image generation, in this work we present a novel method that relies on an information-theoretic alignment measure. In a nutshell, our method uses self-supervised fine-tuning and relies on point-wise mutual information between prompts and images to define a synthetic training set to induce model alignment. Our comparative analysis shows that our method is on-par or superior to the state-of-the-art, yet requires nothing but a pre-trained denoising network to estimate MI and a lightweight fine-tuning strategy.

Data Augmentation (DA) -- enriching training data by adding synthetic samples -- is a technique widely adopted in Computer Vision (CV) and Natural Language Processing (NLP) tasks to improve models performance. Yet, DA has struggled to gain traction in networking contexts, particularly in Traffic Classification (TC) tasks. In this work, we fulfill this gap by benchmarking 18 augmentation functions applied to 3 TC datasets using packet time series as input representation and considering a variety of training conditions. Our results show that (i) DA can reap benefits previously unexplored, (ii) augmentations acting on time series sequence order and masking are better suited for TC than amplitude augmentations and (iii) basic models latent space analysis can help understanding the positive/negative effects of augmentations on classification performance., Comment: to appear at Passive and Active Measurements (PAM), 2024

Data Augmentation (DA)-augmenting training data with synthetic samples-is wildly adopted in Computer Vision (CV) to improve models performance. Conversely, DA has not been yet popularized in networking use cases, including Traffic Classification (TC). In this work, we present a preliminary study of 14 hand-crafted DAs applied on the MIRAGE19 dataset. Our results (i) show that DA can reap benefits previously unexplored in TC and (ii) foster a research agenda on the use of generative models to automate DA design., Comment: to appear at CoNEXT Student Workshop, 2023

Over the last years we witnessed a renewed interest toward Traffic Classification (TC) captivated by the rise of Deep Learning (DL). Yet, the vast majority of TC literature lacks code artifacts, performance assessments across datasets and reference comparisons against Machine Learning (ML) methods. Among those works, a recent study from IMC22 [16] is worth of attention since it adopts recent DL methodologies (namely, few-shot learning, self-supervision via contrastive learning and data augmentation) appealing for networking as they enable to learn from a few samples and transfer across datasets. The main result of [16] on the UCDAVIS19, ISCX-VPN and ISCX-Tor datasets is that, with such DL methodologies, 100 input samples are enough to achieve very high accuracy using an input representation called "flowpic" (i.e., a per-flow 2d histograms of the packets size evolution over time). In this paper (i) we reproduce [16] on the same datasets and (ii) we replicate its most salient aspect (the importance of data augmentation) on three additional public datasets (MIRAGE19, MIRAGE22 and UTMOBILENET21). While we confirm most of the original results, we also found a 20% accuracy drop on some of the investigated scenarios due to a data shift in the original dataset that we uncovered. Additionally, our study validates that the data augmentation strategies studied in [16] perform well on other datasets too. In the spirit of reproducibility and replicability we make all artifacts (code and data) available to the research community at https://tcbenchstack.github.io/tcbench/, Comment: to appear at ACM Internet Traffic Measurement (IMC) 2023, replication track

The popularity of Deep Learning (DL), coupled with network traffic visibility reduction due to the increased adoption of HTTPS, QUIC and DNS-SEC, re-ignited interest towards Traffic Classification (TC). However, to tame the dependency from task-specific large labeled datasets we need to find better ways to learn representations that are valid across tasks. In this work we investigate this problem comparing transfer learning, meta-learning and contrastive learning against reference Machine Learning (ML) tree-based and monolithic DL models (16 methods total). Using two publicly available datasets, namely MIRAGE19 (40 classes) and AppClassNet (500 classes), we show that (i) using large datasets we can obtain more general representations, (ii) contrastive learning is the best methodology and (iii) meta-learning the worst one, and (iv) while ML tree-based cannot handle large tasks but fits well small tasks, by means of reusing learned representations, DL methods are reaching tree-based models performance also for small tasks., Comment: to appear in Traffic Measurements and Analysis (TMA) 2023

While the promises of Multi-Task Learning (MTL) are attractive, characterizing the conditions of its success is still an open problem in Deep Learning. Some tasks may benefit from being learned together while others may be detrimental to one another. From a task perspective, grouping cooperative tasks while separating competing tasks is paramount to reap the benefits of MTL, i.e., reducing training and inference costs. Therefore, estimating task affinity for joint learning is a key endeavor. Recent work suggests that the training conditions themselves have a significant impact on the outcomes of MTL. Yet, the literature is lacking of a benchmark to assess the effectiveness of tasks affinity estimation techniques and their relation with actual MTL performance. In this paper, we take a first step in recovering this gap by (i) defining a set of affinity scores by both revisiting contributions from previous literature as well presenting new ones and (ii) benchmarking them on the Taskonomy dataset. Our empirical campaign reveals how, even in a small-scale scenario, task affinity scoring does not correlate well with actual MTL performance. Yet, some metrics can be more indicative than others., Comment: 7 pages. AAAI'23 - 2nd International Workshop on Practical Deep Learning in the Wild

We show that every quasiconformal contact foliation supports an invariant metric and characterise such foliations by the dynamical property of $C^1$-equicontinuity. We prove that a generalisation of the Weinstein conjecture holds for quasiconformal contact foliations, and provide a lower bound to the number of closed leaves. In particular, we show that the Weinstein conjecture holds for quasiconformal Reeb fields.

Score-based diffusion models are a class of generative models whose dynamics is described by stochastic differential equations that map noise into data. While recent works have started to lay down a theoretical foundation for these models, an analytical understanding of the role of the diffusion time T is still lacking. Current best practice advocates for a large T to ensure that the forward dynamics brings the diffusion sufficiently close to a known and simple noise distribution; however, a smaller value of T should be preferred for a better approximation of the score-matching objective and higher computational efficiency. Starting from a variational interpretation of diffusion models, in this work we quantify this trade-off, and suggest a new method to improve quality and efficiency of both training and sampling, by adopting smaller diffusion times. Indeed, we show how an auxiliary model can be used to bridge the gap between the ideal and the simulated forward dynamics, followed by a standard reverse diffusion process. Empirical results support our analysis; for image data, our method is competitive w.r.t. the state-of-the-art, according to standard sample quality metrics and log-likelihood.

The $n$-torus is the the unique closed manifold supporting a set of $n$ linearly independent closed $1$-forms. In this paper we improve on this result and show that the torus is the unique closed $n$-dimensional manifold supporting a linearly independent set consisting of $(n-1)$ closed $1$-forms whose product determines a non-zero cohomological class., Comment: arXiv admin note: text overlap with arXiv:2205.14190

We consider contact foliations: objects which generalise to higher dimensions the flow of the Reeb vector field on contact manifolds. We list a number of properties of such foliations, and propose two conjectures about the topological types of their leaves, both of which coincide with the classical Weinstein conjecture in the case of contact flows. We give positive partial results for our conjectures in particular cases -- when the holonomy of the contact foliation preserves a Riemannian metric, for instance -- extending already established results from the field of Contact Dynamics.

Sentiment Analysis is a field of Natural Language Processing which aims at classifying the author's sentiment in text. This paper first describes a sentiment analysis model for students' comments about professor performance. The model achieved impressive results for comments collected from student surveys conducted at a private university in 2019/20. Then, it applies the model to different scenarios: (1) in-person classes taught in 2019 (pre-COVID); (2) the emergency shift to online, synchronous classes taught in the first semester of 2020 (early-COVID); and (3) the planned online classes taught in the second semester of 2020 (late-COVID). The results show that students acknowledged the effort professors did to keep classes running during the first semester of 2020, and that the enthusiasm continued throughout the second semester. Furthermore, the results show that students evaluated professors' performance for online courses better than for in-person courses. [For the full proceedings, see ED615472.]

Ongoing traffic changes, including those triggered by the COVID-19 pandemic, reveal the necessity to adapt our public transport systems to the ever-changing users' needs. This work shows that single and multi objective stances can be synergistically combined to better answer the transit network design problem (TNDP). Single objective formulations are dynamically inferred from the rating of networks in the approximated (multi-objective) Pareto Front, where a regression approach is used to infer the optimal weights of transfer needs, times, distances, coverage, and costs. As a guiding case study, the solution is applied to the multimodal public transport network in the city of Lisbon, Portugal. The system takes individual trip data given by smartcard validations at CARRIS buses and METRO subway stations and uses them to estimate the origin-destination demand in the city. Then, Genetic Algorithms are used, considering both single and multi objective approaches, to redesign the bus network that better fits the observed traffic demand. The proposed TNDP optimization proved to improve results, with reductions in objective functions of up to 28.3%. The system managed to extensively reduce the number of routes, and all passenger related objectives, including travel time and transfers per trip, significantly improve. Grounded on automated fare collection data, the system can incrementally redesign the bus network to dynamically handle ongoing changes to the city traffic.

While Deep Learning (DL) technologies are a promising tool to solve networking problems that map to classification tasks, their computational complexity is still too high with respect to real-time traffic measurements requirements. To reduce the DL inference cost, we propose a novel caching paradigm, that we named approximate-key caching, which returns approximate results for lookups of selected input based on cached DL inference results. While approximate cache hits alleviate DL inference workload and increase the system throughput, they however introduce an approximation error. As such, we couple approximate-key caching with an error-correction principled algorithm, that we named auto-refresh. We analytically model our caching system performance for classic LRU and ideal caches, we perform a trace-driven evaluation of the expected performance, and we compare the benefits of our proposed approach with the state-of-the-art similarity caching -- testifying the practical interest of our proposal., Comment: Accepted at IEEE Infocom 2022

The recent popularity growth of Deep Learning (DL) re-ignited the interest towards traffic classification, with several studies demonstrating the accuracy of DL-based classifiers to identify Internet applications' traffic. Even with the aid of hardware accelerators (GPUs, TPUs), DL model training remains expensive, and limits the ability to operate frequent model updates necessary to fit to the ever evolving nature of Internet traffic, and mobile traffic in particular. To address this pain point, in this work we explore Incremental Learning (IL) techniques to add new classes to models without a full retraining, hence speeding up model's updates cycle. We consider iCarl, a state of the art IL method, and MIRAGE-2019, a public dataset with traffic from 40 Android apps, aiming to understand "if there is a case for incremental learning in traffic classification". By dissecting iCarl internals, we discuss ways to improve its design, contributing a revised version, namely iCarl+. Despite our analysis reveals their infancy, IL techniques are a promising research area on the roadmap towards automated DL-based traffic analysis systems., Comment: Accepted for publication at Network Traffic Measurement and Analysis Conference (TMA), September 2021

Live traffic analysis at the first aggregation point in the ISP network enables the implementation of complex traffic engineering policies but is limited by the scarce processing capabilities, especially for Deep Learning (DL) based analytics. The introduction of specialized hardware accelerators i.e., Tensor Processing Unit (TPU), offers the opportunity to enhance the processing capabilities of network devices at the edge. Yet, to date, no packet processing pipeline is capable of offering DL-based analysis capabilities in the data-plane, without interfering with network operations. In this paper, we present FENXI, a system to run complex analytics by leveraging TPU. The design of FENXI decouples forwarding operations and traffic analytics which operates at different granularities i.e., packet and flow levels. We conceive two independent modules that asynchronously communicate to exchange network data and analytics results, and design data structures to extract flow level statistics without impacting per-packet processing. We prototyped and evaluated FENXI on general-purpose servers considering both adversarial and realistic network conditions. Our analysis shows that FENXI can sustain 100 Gbps line rate traffic processing requiring only limited resources, while also dynamically adapting to variable network conditions., Comment: 14 pages, 12 figures. Accepted for publication at the Sixth ACM/IEEE Symposium on Edge Computing (SEC'21), December 2021

Bike sharing demand is increasing in large cities worldwide. The proper functioning of bike-sharing systems is, nevertheless, dependent on a balanced geographical distribution of bicycles throughout a day. In this context, understanding the spatiotemporal distribution of check-ins and check-outs is key for station balancing and bike relocation initiatives. Still, recent contributions from deep learning and distance-based predictors show limited success on forecasting bike sharing demand. This consistent observation is hypothesized to be driven by: i) the strong dependence between demand and the meteorological and situational context of stations; and ii) the absence of spatial awareness as most predictors are unable to model the effects of high-low station load on nearby stations. This work proposes a comprehensive set of new principles to incorporate both historical and prospective sources of spatial, meteorological, situational and calendrical context in predictive models of station demand. To this end, a new recurrent neural network layering composed by serial long-short term memory (LSTM) components is proposed with two major contributions: i) the feeding of multivariate time series masks produced from historical context data at the input layer, and ii) the time-dependent regularization of the forecasted time series using prospective context data. This work further assesses the impact of incorporating different sources of context, showing the relevance of the proposed principles for the community even though not all improvements from the context-aware predictors yield statistical significance.

The increasing success of Machine Learning (ML) and Deep Learning (DL) has recently re-sparked interest towards traffic classification. While classification of known traffic is a well investigated subject with supervised classification tools (such as ML and DL models) are known to provide satisfactory performance, detection of unknown (or zero-day) traffic is more challenging and typically handled by unsupervised techniques (such as clustering algorithms). In this paper, we share our experience on a commercial-grade DL traffic classification engine that is able to (i) identify known applications from encrypted traffic, as well as (ii) handle unknown zero-day applications. In particular, our contribution for (i) is to perform a thorough assessment of state of the art traffic classifiers in commercial-grade settings comprising few thousands of very fine grained application labels, as opposite to the few tens of classes generally targeted in academic evaluations. Additionally, we contribute to the problem of (ii) detection of zero-day applications by proposing a novel technique, tailored for DL models, that is significantly more accurate and light-weight than the state of the art. Summarizing our main findings, we gather that (i) while ML and DL models are both equally able to provide satisfactory solution for classification of known traffic, however (ii) the non-linear feature extraction process of the DL backbone provides sizeable advantages for the detection of unknown classes.

The closed design of mobile devices -- with the increased security and consistent user interfaces -- is in large part responsible for their becoming the dominant platform for accessing the Internet. These benefits, however, are not without a cost. Their operation of mobile devices and their apps is not easy to understand by either users or operators. We argue for recovering transparency and control on mobile devices through an extensible platform that can intercept and modify traffic before leaving the device or, on arrival, before it reaches the operating system. Conceptually, this is the same view of the traffic that a traditional middlebox would have at the far end of the first link in the network path. We call this platform ``middlebox zero'' or MBZ. By being on-board, MBZ also leverages local context as it processes the traffic and complements the network-wide view of standard middleboxes. We discuss the challenges of the MBZ approach, sketch a working design, and illustrate its potential with some concrete examples., Comment: The paper is a position piece under review

Support for things roaming internationally has become critical for Internet of Things (IoT) verticals, from connected cars to smart meters and wearables, and explains the commercial success of Machine-to-Machine (M2M) platforms. We analyze IoT verticals operating with connectivity via IoT SIMs, and present the first large-scale study of commercially deployed IoT SIMs for energy meters. We also present the first characterization of an operational M2M platform and the first analysis of the rather opaque associated ecosystem. For operators, the exponential growth of IoT has meant increased stress on the infrastructure shared with traditional roaming traffic. Our analysis quantifies the adoption of roaming by M2M platforms and the impact they have on the underlying visited Mobile Network Operators (MNOs). To manage the impact of massive deployments of device operating with an IoT SIM, operators must be able to distinguish between the latter and traditional inbound roamers. We build a comprehensive dataset capturing the device population of a large European MNO over three weeks. With this, we propose and validate a classification approach that can allow operators to distinguish inbound roaming IoT devices.

Critical Path Analysis (CPA) studies the delivery of webpages to identify page resources, their interrelations, as well as their impact on the page loading latency. Despite CPA being a generic methodology, its mechanisms have been applied only to browsers and web traffic, but those do not directly apply to study generic mobile apps. Likewise, web browsing represents only a small fraction of the overall mobile traffic. In this paper, we take a first step towards filling this gap by exploring how CPA can be performed for generic mobile applications. We propose Mobile Critical Path Analysis (MCPA), a methodology based on passive and active network measurements that is applicable to a broad set of apps to expose a fine-grained view of their traffic dynamics. We validate MCPA on popular apps across different categories and usage scenarios. We show that MCPA can identify user interactions with mobile apps only based on traffic monitoring, and the relevant network activities that are bottlenecks. Overall, we observe that apps spend 60% of time and 84% of bytes on critical traffic on average, corresponding to +22% time and +13% bytes than what observed for browsing.