arXiv subjects

Fabrizio Silvestri

Publications and source records attributed to Fabrizio Silvestri.

103 records · Page 6Linked to original sources

arXiv2020

Preserving Integrity in Online Social Networks

Online social networks provide a platform for sharing information and free expression. However, these networks are also used for malicious purposes, such as distributing misinformation and hate speech, selling illegal drugs, and coordinating sex trafficking or child exploitation. This paper surveys the state of the art in keeping online platforms and their users safe from such harm, also known as the problem of preserving integrity. This survey comes from the perspective of having to combat a broad spectrum of integrity violations at Facebook. We highlight the techniques that have been proven useful in practice and that deserve additional attention from the academic community. Instead of discussing the many individual violation types, we identify key aspects of the social-media eco-system, each of which is common to a wide variety violation types. Furthermore, each of these components represents an area for research and development, and the innovations that are found can be applied widely.

cs.SI↗

arXiv2020

Neighborhood Sensitive Mapping for Zero-Shot Classification using Independently Learned Semantic Embeddings

In a traditional setting, classifiers are trained to approximate a target function $f:X \rightarrow Y$ where at least a sample for each $y \in Y$ is presented to the training algorithm. In a zero-shot setting we have a subset of the labels $\hat{Y} \subset Y$ for which we do not observe any corresponding training instance. Still, the function $f$ that we train must be able to correctly assign labels also on $\hat{Y}$. In practice, zero-shot problems are very important especially when the label set is large and the cost of editorially label samples for all possible values in the label set might be prohibitively high. Most recent approaches to zero-shot learning are based on finding and exploiting relationships between labels using semantic embeddings. We show in this paper that semantic embeddings, despite being very good at capturing relationships between labels, are not very good at capturing the relationships among labels in a data-dependent manner. For this reason, we propose a novel two-step process for learning a zero-shot classifier. In the first step, we learn what we call a \emph{property embedding space} capturing the "\emph{learnable}" features of the label set. Then, we exploit the learned properties in order to reduce the generalization error for a linear nearest neighbor-based classifier.

cs.LG↗

arXiv2020

Concept Matching for Low-Resource Classification

We propose a model to tackle classification tasks in the presence of very little training data. To this aim, we approximate the notion of exact match with a theoretically sound mechanism that computes a probability of matching in the input space. Importantly, the model learns to focus on elements of the input that are relevant for the task at hand; by leveraging highlighted portions of the training data, an error boosting technique guides the learning process. In practice, it increases the error associated with relevant parts of the input by a given factor. Remarkable results on text classification tasks confirm the benefits of the proposed approach in both balanced and unbalanced cases, thus being of practical use when labeling new examples is expensive. In addition, by inspecting its weights, it is often possible to gather insights on what the model has learned.

cs.LG↗

arXiv2020

Nanostructure-modulated planar high spectral resolution spectro-polarimeter

We present a planar spectro-polarimeter based on Fabry-P{é}rot cavities with embedded polarization-sensitive high-index nanostructures. A $7~μ$m-thick spectro-polarimetric system for 3 spectral bands and 2 linear polarization states is experimentally demonstrated. Furthermore, an optimal design is theoretically proposed, estimating that a system with a bandwidth of 127~nm and a spectral resolution of 1~nm is able to reconstruct the first three Stokes parameters \textcolor{black}{with a signal-to-noise ratio of -13.14~dB with respect to the the shot noise limited SNR}. The pixelated spectro-polarimetric system can be directly integrated on a sensor, thus enabling applicability in a variety of miniaturized optical devices, including but not limited to satellites for Earth observation.

physics.optics↗

arXiv2020

Color filter arrays based on dielectric metasurface elements

Digital imaging has been steadily improving over the past decades and we are moving towards a wide use of multi- and hyperspectral cameras. A key component of such imaging systems are color filter arrays, which define the spectrum of light detected by each camera pixel. Hence, it is essential to develop a variable, robust and scalable way for controlling the transmission of light. Nanostructured surfaces, also known as metasurfaces, offer a promising solution as their transmission spectra can be controlled by shaping the wavelength-dependent scattering properties of their constituting elements. Here we present, metasurfaces based on silicon nanodisks, which provide filter functions with amplitudes reaching 70-90% of transmission, and well suitable for RGB and CMY color filter arrays, the initial stage towards the further development of hyperspectral filters. We suggest and discuss possible ways to expand the color gamut and improve the color values of such optical filters.

physics.optics↗

arXiv2019

Misspelling Oblivious Word Embeddings

In this paper we present a method to learn word embeddings that are resilient to misspellings. Existing word embeddings have limited applicability to malformed texts, which contain a non-negligible amount of out-of-vocabulary words. We propose a method combining FastText with subwords and a supervised task of learning misspelling patterns. In our method, misspellings of each word are embedded close to their correct variants. We train these embeddings on a new dataset we are releasing publicly. Finally, we experimentally show the advantages of this approach on both intrinsic and extrinsic NLP tasks using public test sets.

cs.CL↗

arXiv2019

Reflection confocal nanoscopy using a super-oscillatory lens

A Superoscillatory lens (SOL) is known to produce a sub-diffraction hotspot which is useful for high-resolution imaging. However, high-energy rings called sidelobes coexist with the central hotspot. Additionally, SOLs have not yet been directly used to image reflective objects due to low efficiency and poor imaging properties. We propose a novel reflection confocal nanoscope which mitigates these issues by relaying the SOL intensity pattern onto the object and use conventional optics for detection. We experimentally demonstrate super-resolution by imaging double bars with 330 nm separation using a 632.8 nm excitation and a 0.95 NA objective. We also discuss the enhanced contrast properties of the SOL nanoscope against a laser confocal microscope, and the degradation of performance while imaging large objects.

physics.optics↗

arXiv2017

Interpretable Predictions of Tree-based Ensembles via Actionable Feature Tweaking

Machine-learned models are often described as "black boxes". In many real-world applications however, models may have to sacrifice predictive power in favour of human-interpretability. When this is the case, feature engineering becomes a crucial task, which requires significant and time-consuming human effort. Whilst some features are inherently static, representing properties that cannot be influenced (e.g., the age of an individual), others capture characteristics that could be adjusted (e.g., the daily amount of carbohydrates taken). Nonetheless, once a model is learned from the data, each prediction it makes on new instances is irreversible - assuming every instance to be a static point located in the chosen feature space. There are many circumstances however where it is important to understand (i) why a model outputs a certain prediction on a given instance, (ii) which adjustable features of that instance should be modified, and finally (iii) how to alter such a prediction when the mutated instance is input back to the model. In this paper, we present a technique that exploits the internals of a tree-based ensemble classifier to offer recommendations for transforming true negative instances into positively predicted ones. We demonstrate the validity of our approach using an online advertising application. First, we design a Random Forest classifier that effectively separates between two types of ads: low (negative) and high (positive) quality ads (instances). Then, we introduce an algorithm that provides recommendations that aim to transform a low quality ad (negative instance) into a high quality one (positive instance). Finally, we evaluate our approach on a subset of the active inventory of a large ad network, Yahoo Gemini.

stat.ML↗

arXiv2017

Pornography consumption in Social Media

The structure of a social network is fundamentally related to the interests of its members. People assort spontaneously based on the topics that are relevant to them, forming social groups that revolve around different subjects. Online social media are also favorable ecosystems for the formation of topical communities centered on matters that are not commonly taken up by the general public because of the embarrassment, discomfort, or shock they may cause. Those are communities that depict or discuss what are usually referred to as deviant behaviors, conducts that are commonly considered inappropriate because they are somehow violative of society's norms or moral standards that are shared among the majority of the members of society. Pornography consumption, drug use, excessive drinking, illegal hunting, eating disorders, or any self-harming or addictive practice are all examples of deviant behaviors.

cs.CY↗

arXiv2016

On the Behaviour of Deviant Communities in Online Social Networks

On-line social networks are complex ensembles of inter-linked communities that interact on different topics. Some communities are characterized by what are usually referred to as deviant behaviors, conducts that are commonly considered inappropriate with respect to the society's norms or moral standards. Eating disorders, drug use, and adult content consumption are just a few examples. We refer to such communities as deviant networks. It is commonly believed that such deviant networks are niche, isolated social groups, whose activity is well separated from the mainstream social-media life. According to this assumption, research studies have mostly considered them in isolation. In this work we focused on adult content consumption networks, which are present in many on-line social media and in the Web in general. We found that few small and densely connected communities are responsible for most of the content production. Differently from previous work, we studied how such communities interact with the whole social network. We found that the produced content flows to the rest of the network mostly directly or through bridge-communities, reaching at least 450 times more users. We also show that a large fraction of the users can be inadvertently exposed to such content through indirect content resharing. We also discuss a demographic analysis of the producers and consumers networks. Finally, we show that it is easily possible to identify a few core users to radically uproot the diffusion process. We aim at setting the basis to study deviant communities in context.

cs.SI↗

arXiv2016

Scalable Semantic Matching of Queries to Ads in Sponsored Search Advertising

Sponsored search represents a major source of revenue for web search engines. This popular advertising model brings a unique possibility for advertisers to target users' immediate intent communicated through a search query, usually by displaying their ads alongside organic search results for queries deemed relevant to their products or services. However, due to a large number of unique queries it is challenging for advertisers to identify all such relevant queries. For this reason search engines often provide a service of advanced matching, which automatically finds additional relevant queries for advertisers to bid on. We present a novel advanced matching approach based on the idea of semantic embeddings of queries and ads. The embeddings were learned using a large data set of user search sessions, consisting of search queries, clicked ads and search links, while utilizing contextual information such as dwell time and skipped ads. To address the large-scale nature of our problem, both in terms of data and vocabulary size, we propose a novel distributed algorithm for training of the embeddings. Finally, we present an approach for overcoming a cold-start problem associated with new ads and queries. We report results of editorial evaluation and online tests on actual search traffic. The results show that our approach significantly outperforms baselines in terms of relevance, coverage, and incremental revenue. Lastly, we open-source learned query embeddings to be used by researchers in computational advertising and related fields.

cs.IR↗

arXiv2011

Efficient Diversification of Web Search Results

In this paper we analyze the efficiency of various search results diversification methods. While efficacy of diversification approaches has been deeply investigated in the past, response time and scalability issues have been rarely addressed. A unified framework for studying performance and feasibility of result diversification solutions is thus proposed. First we define a new methodology for detecting when, and how, query results need to be diversified. To this purpose, we rely on the concept of "query refinement" to estimate the probability of a query to be ambiguous. Then, relying on this novel ambiguity detection method, we deploy and compare on a standard test set, three different diversification methods: IASelect, xQuAD, and OptSelect. While the first two are recent state-of-the-art proposals, the latter is an original algorithm introduced in this paper. We evaluate both the efficiency and the effectiveness of our approach against its competitors by using the standard TREC Web diversification track testbed. Results shown that OptSelect is able to run two orders of magnitude faster than the two other state-of-the-art approaches and to obtain comparable figures in diversification effectiveness.

cs.IR↗

arXiv2004

Design of a Parallel and Distributed Web Search Engine

This paper describes the architecture of MOSE (My Own Search Engine), a scalable parallel and distributed engine for searching the web. MOSE was specifically designed to efficiently exploit affordable parallel architectures, such as clusters of workstations. Its modular and scalable architecture can easily be tuned to fulfill the bandwidth requirements of the application at hand. Both task-parallel and data-parallel approaches are exploited within MOSE in order to increase the throughput and efficiently use communication, storing and computational resources. We used a collection of html documents as a benchmark, and conducted preliminary experiments on a cluster of three SMP Linux PCs.

cs.IR↗