Search

Digital platforms such as social media and e-commerce websites adopt Recommender Systems to provide value to the user. However, the social consequences deriving from their adoption are still unclear. Many scholars argue that recommenders may lead to detrimental effects, such as bias-amplification deriving from the feedback loop between algorithmic suggestions and users' choices. Nonetheless, the extent to which recommenders influence changes in users leaning remains uncertain. In this context, it is important to provide a controlled environment for evaluating the recommendation algorithm before deployment. To address this, we propose a stochastic simulation framework that mimics user-recommender system interactions in a long-term scenario. In particular, we simulate the user choices by formalizing a user model, which comprises behavioral aspects, such as the user resistance towards the recommendation algorithm and their inertia in relying on the received suggestions. Additionally, we introduce two novel metrics for quantifying the algorithm's impact on user preferences, specifically in terms of drift over time. We conduct an extensive evaluation on multiple synthetic datasets, aiming at testing the robustness of our framework when considering different scenarios and hyper-parameters setting. The experimental results prove that the proposed methodology is effective in detecting and quantifying the drift over the users preferences by means of the simulation. All the code and data used to perform the experiments are publicly available.

Recent developments in recommendation have harnessed the collaborative power of graph neural networks (GNNs) in learning users' preferences from user-item networks. Despite emerging regulations addressing fairness of automated systems, unfairness issues in graph collaborative filtering remain underexplored, especially from the consumer's perspective. Despite numerous contributions on consumer unfairness, only a few of these works have delved into GNNs. A notable gap exists in the formalization of the latest mitigation algorithms, as well as in their effectiveness and reliability on cutting-edge models. This paper serves as a solid response to recent research highlighting unfairness issues in graph collaborative filtering by reproducing one of the latest mitigation methods. The reproduced technique adjusts the system fairness level by learning a fair graph augmentation. Under an experimental setup based on 11 GNNs, 5 non-GNN models, and 5 real-world networks across diverse domains, our investigation reveals that fair graph augmentation is consistently effective on high-utility models and large datasets. Experiments on the transferability of the fair augmented graph open new issues for future recommendation studies. Source code: https://github.com/jackmedda/FA4GCF.

Slate recommendation is a technique commonly used on streaming platforms and e-commerce sites to present multiple items together. A significant challenge with slate recommendation is managing the complex combinatorial choice space. Traditional methods often simplify this problem by assuming users engage with only one item at a time. However, this simplification does not reflect the reality, as users often interact with multiple items simultaneously. In this paper, we address the general slate recommendation problem, which accounts for simultaneous engagement with multiple items. We propose a generative approach using Diffusion Models, leveraging their ability to learn structures in high-dimensional data. Our model generates high-quality slates that maximize user satisfaction by overcoming the challenges of the combinatorial choice space. Furthermore, our approach enhances the diversity of recommendations. Extensive offline evaluations on applications such as music playlist generation and e-commerce bundle recommendations show that our model outperforms state-of-the-art baselines in both relevance and diversity., Comment: 9 pages, 5 figures, 3 tables

We represent preferences that exhibit absolute or relative attitudes towards ambiguity without assuming convexity of preferences. Our analysis is motivated by the recent experimental evidence by Baillon and Placido (2019) indicating that ambiguity becomes more tolerable as individuals are better off overall. Decreasing absolute ambiguity aversion is characterized by constant superadditive certainty equivalents and admits an act-dependent variational representation (Maccheroni et al., 2006). Decreasing relative ambiguity aversion relates to positive superhomogeneity and admits an act-dependent confidence preference representation (Chateauneuf and Faro, 2009). We apply our characterizations to retrieve a classic risk sharing result on the efficiency of trade and subjective beliefs of the individuals (Rigotti et al., 2008).

In the realm of personalization, integrating diverse information sources such as consumption signals and content-based representations is becoming increasingly critical to build state-of-the-art solutions. In this regard, two of the biggest trends in research around this subject are Graph Neural Networks (GNNs) and Foundation Models (FMs). While GNNs emerged as a popular solution in industry for powering personalization at scale, FMs have only recently caught attention for their promising performance in personalization tasks like ranking and retrieval. In this paper, we present a graph-based foundation modeling approach tailored to personalization. Central to this approach is a Heterogeneous GNN (HGNN) designed to capture multi-hop content and consumption relationships across a range of recommendable item types. To ensure the generality required from a Foundation Model, we employ a Large Language Model (LLM) text-based featurization of nodes that accommodates all item types, and construct the graph using co-interaction signals, which inherently transcend content specificity. To facilitate practical generalization, we further couple the HGNN with an adaptation mechanism based on a two-tower (2T) architecture, which also operates agnostically to content type. This multi-stage approach ensures high scalability; while the HGNN produces general purpose embeddings, the 2T component models in a continuous space the sheer size of user-item interaction data. Our comprehensive approach has been rigorously tested and proven effective in delivering recommendations across a diverse array of products within a real-world, industrial audio streaming platform.

In the ever-evolving digital audio landscape, Spotify, well-known for its music and talk content, has recently introduced audiobooks to its vast user base. While promising, this move presents significant challenges for personalized recommendations. Unlike music and podcasts, audiobooks, initially available for a fee, cannot be easily skimmed before purchase, posing higher stakes for the relevance of recommendations. Furthermore, introducing a new content type into an existing platform confronts extreme data sparsity, as most users are unfamiliar with this new content type. Lastly, recommending content to millions of users requires the model to react fast and be scalable. To address these challenges, we leverage podcast and music user preferences and introduce 2T-HGNN, a scalable recommendation system comprising Heterogeneous Graph Neural Networks (HGNNs) and a Two Tower (2T) model. This novel approach uncovers nuanced item relationships while ensuring low latency and complexity. We decouple users from the HGNN graph and propose an innovative multi-link neighbor sampler. These choices, together with the 2T component, significantly reduce the complexity of the HGNN model. Empirical evaluations involving millions of users show significant improvement in the quality of personalized recommendations, resulting in a +46% increase in new audiobooks start rate and a +23% boost in streaming rates. Intriguingly, our model's impact extends beyond audiobooks, benefiting established products like podcasts., Comment: To appear in The Web Conference 2024 proceedings

Efforts in the recommendation community are shifting from the sole emphasis on utility to considering beyond-utility factors, such as fairness and robustness. Robustness of recommendation models is typically linked to their ability to maintain the original utility when subjected to attacks. Limited research has explored the robustness of a recommendation model in terms of fairness, e.g., the parity in performance across groups, under attack scenarios. In this paper, we aim to assess the robustness of graph-based recommender systems concerning fairness, when exposed to attacks based on edge-level perturbations. To this end, we considered four different fairness operationalizations, including both consumer and provider perspectives. Experiments on three datasets shed light on the impact of perturbations on the targeted fairness notion, uncovering key shortcomings in existing evaluation protocols for robustness. As an example, we observed perturbations affect consumer fairness on a higher extent than provider fairness, with alarming unfairness for the former. Source code: https://github.com/jackmedda/CPFairRobust

Federated Learning (FL) has emerged as a key approach for distributed machine learning, enhancing online personalization while ensuring user data privacy. Instead of sending private data to a central server as in traditional approaches, FL decentralizes computations: devices train locally and share updates with a global server. A primary challenge in this setting is achieving fast and accurate model training - vital for recommendation systems where delays can compromise user engagement. This paper introduces FedFNN, an algorithm that accelerates decentralized model training. In FL, only a subset of users are involved in each training epoch. FedFNN employs supervised learning to predict weight updates from unsampled users, using updates from the sampled set. Our evaluations, using real and synthetic data, show: 1. FedFNN achieves training speeds 5x faster than leading methods, maintaining or improving accuracy; 2. the algorithm's performance is consistent regardless of client cluster variations; 3. FedFNN outperforms other methods in scenarios with limited client availability, converging more quickly.

In recommendation literature, explainability and fairness are becoming two prominent perspectives to consider. However, prior works have mostly addressed them separately, for instance by explaining to consumers why a certain item was recommended or mitigating disparate impacts in recommendation utility. None of them has leveraged explainability techniques to inform unfairness mitigation. In this paper, we propose an approach that relies on counterfactual explanations to augment the set of user-item interactions, such that using them while inferring recommendations leads to fairer outcomes. Modeling user-item interactions as a bipartite graph, our approach augments the latter by identifying new user-item edges that not only can explain the original unfairness by design, but can also mitigate it. Experiments on two public data sets show that our approach effectively leads to a better trade-off between fairness and recommendation utility compared with state-of-the-art mitigation procedures. We further analyze the characteristics of added edges to highlight key unfairness patterns. Source code available at https://github.com/jackmedda/RS-BGExplainer/tree/cikm2023., Comment: Accepted as a short paper at CIKM 2023

Gravitational wave astronomy pipelines rely on template waveform models for searches and parameter estimation purposes. For coalescing binary neutron stars (BNS), such models need to accurately reproduce numerical relativity (NR) up to merger, in order to provide robust estimate of the stars' equation of state - dependent parameters. In this work we present an improved version of the Effective One Body (EOB) model $\tt TEOBResumS$ for gravitational waves from BNS systems. Building upon recent post-Newtonian calculations, we include subleading order tidal terms in the waveform multipoles and EOB metric potentials, as well as add up to 5.5PN terms in the gyro-gravitomagnetic functions entering the spin-orbit sector of the model. In order to further improve the EOB-NR agreement in the last few orbital cycles before merger, we introduce next-to-quasicircular corrections in the waveform -- informed by a large number of BNS NR simulations -- and introduce a new NR-informed parameter entering the tidal sector of our conservative dynamics. The performance of our model is then validated against 14 new eccentricity reduced simulations of unequal mass, spinning binaries with varying equation of state. A time-domain phasing analysis and mismatch computations demonstrate that the new model overall improves over the previous version of $\tt TEOBResumS$. Finally, we present a closed-form frequency domain representation of the (tidal) amplitude and phase of the new model. This representation accounts for mass-ratio, aligned spin and (resummed) spin-quadrupole effects in the tidal phase and -- within the calibration region -- it is faithful to the original model., Comment: 17 pages, 6 figures

Nowadays, research into personalization has been focusing on explainability and fairness. Several approaches proposed in recent works are able to explain individual recommendations in a post-hoc manner or by explanation paths. However, explainability techniques applied to unfairness in recommendation have been limited to finding user/item features mostly related to biased recommendations. In this paper, we devised a novel algorithm that leverages counterfactuality methods to discover user unfairness explanations in the form of user-item interactions. In our counterfactual framework, interactions are represented as edges in a bipartite graph, with users and items as nodes. Our bipartite graph explainer perturbs the topological structure to find an altered version that minimizes the disparity in utility between the protected and unprotected demographic groups. Experiments on four real-world graphs coming from various domains showed that our method can systematically explain user unfairness on three state-of-the-art GNN-based recommendation models. Moreover, an empirical evaluation of the perturbed network uncovered relevant patterns that justify the nature of the unfairness discovered by the generated explanations. The source code and the preprocessed data sets are available at https://github.com/jackmedda/RS-BGExplainer.

The proliferation of harmful content shared online poses a threat to online information integrity and the integrity of discussion across platforms. Despite various moderation interventions adopted by social media platforms, researchers and policymakers are calling for holistic solutions. This study explores how a target platform could leverage content that has been deemed harmful on a source platform by investigating the behavior and characteristics of Twitter users responsible for sharing moderated YouTube videos. Using a large-scale dataset of 600M tweets related to the 2020 U.S. election, we find that moderated Youtube videos are extensively shared on Twitter and that users who share these videos also endorse extreme and conspiratorial ideologies. A fraction of these users are eventually suspended by Twitter, but they do not appear to be involved in state-backed information operations. The findings of this study highlight the complex and interconnected nature of harmful cross-platform information diffusion, raising the need for cross-platform moderation strategies., Comment: 14 pages, 8 figures

Diversity maximization aims to select a diverse and representative subset of items from a large dataset. It is a fundamental optimization task that finds applications in data summarization, feature selection, web search, recommender systems, and elsewhere. However, in a setting where data items are associated with different groups according to sensitive attributes like sex or race, it is possible that algorithmic solutions for this task, if left unchecked, will under- or over-represent some of the groups. Therefore, we are motivated to address the problem of \emph{max-min diversification with fairness constraints}, aiming to select $k$ items to maximize the minimum distance between any pair of selected items while ensuring that the number of items selected from each group falls within predefined lower and upper bounds. In this work, we propose an exact algorithm based on integer linear programming that is suitable for small datasets as well as a $\frac{1-\varepsilon}{5}$-approximation algorithm for any $\varepsilon \in (0, 1)$ that scales to large datasets. Extensive experiments on real-world datasets demonstrate the superior performance of our proposed algorithms over existing ones., Comment: 13 pages, 8 figures, to appear in SDM '23

Diversity maximization is a fundamental problem with wide applications in data summarization, web search, and recommender systems. Given a set $X$ of $n$ elements, it asks to select a subset $S$ of $k \ll n$ elements with maximum \emph{diversity}, as quantified by the dissimilarities among the elements in $S$. In this paper, we focus on the diversity maximization problem with fairness constraints in the streaming setting. Specifically, we consider the max-min diversity objective, which selects a subset $S$ that maximizes the minimum distance (dissimilarity) between any pair of distinct elements within it. Assuming that the set $X$ is partitioned into $m$ disjoint groups by some sensitive attribute, e.g., sex or race, ensuring \emph{fairness} requires that the selected subset $S$ contains $k_i$ elements from each group $i \in [1,m]$. A streaming algorithm should process $X$ sequentially in one pass and return a subset with maximum \emph{diversity} while guaranteeing the fairness constraint. Although diversity maximization has been extensively studied, the only known algorithms that can work with the max-min diversity objective and fairness constraints are very inefficient for data streams. Since diversity maximization is NP-hard in general, we propose two approximation algorithms for fair diversity maximization in data streams, the first of which is $\frac{1-\varepsilon}{4}$-approximate and specific for $m=2$, where $\varepsilon \in (0,1)$, and the second of which achieves a $\frac{1-\varepsilon}{3m+2}$-approximation for an arbitrary $m$. Experimental results on real-world and synthetic datasets show that both algorithms provide solutions of comparable quality to the state-of-the-art algorithms while running several orders of magnitude faster in the streaming setting., Comment: 13 pages, 11 figures; published in ICDE 2022

Recommender systems typically suggest to users content similar to what they consumed in the past. If a user happens to be exposed to strongly polarized content, she might subsequently receive recommendations which may steer her towards more and more radicalized content, eventually being trapped in what we call a "radicalization pathway". In this paper, we study the problem of mitigating radicalization pathways using a graph-based approach. Specifically, we model the set of recommendations of a "what-to-watch-next" recommender as a d-regular directed graph where nodes correspond to content items, links to recommendations, and paths to possible user sessions. We measure the "segregation" score of a node representing radicalized content as the expected length of a random walk from that node to any node representing non-radicalized content. High segregation scores are associated to larger chances to get users trapped in radicalization pathways. Hence, we define the problem of reducing the prevalence of radicalization pathways by selecting a small number of edges to "rewire", so to minimize the maximum of segregation scores among all radicalized nodes, while maintaining the relevance of the recommendations. We prove that the problem of finding the optimal set of recommendations to rewire is NP-hard and NP-hard to approximate within any factor. Therefore, we turn our attention to heuristics, and propose an efficient yet effective greedy algorithm based on the absorbing random walk theory. Our experiments on real-world datasets in the context of video and news recommendations confirm the effectiveness of our proposal., Comment: To appear in the Web conference 2022 (WWW '22)

People recommender systems may affect the exposure that users receive in social networking platforms, influencing attention dynamics and potentially strengthening pre-existing inequalities that disproportionately affect certain groups. In this paper we introduce a model to simulate the feedback loop created by multiple rounds of interactions between users and a link recommender in a social network. This allows us to study the long-term consequences of those particular recommendation algorithms. Our model is equipped with several parameters to control (i) the level of homophily in the network, (ii) the relative size of the groups, (iii) the choice among several state-of-the-art link recommenders, and (iv) the choice among three different user behavior models, that decide which recommendations are accepted or rejected. Our extensive experimentation with the proposed model shows that a minority group, if homophilic enough, can get a disproportionate advantage in exposure from all link recommenders. Instead, when it is heterophilic, it gets under-exposed. Moreover, while the homophily level of the minority affects the speed of the growth of the disparate exposure, the relative size of the minority affects the magnitude of the effect. Finally, link recommenders strengthen exposure inequalities at the individual level, exacerbating the "rich-get-richer" effect: this happens for both the minority and the majority class and independently of their level of homophily., Comment: To appear in ICWSM 2022

Numerical studies of the dynamics of gravitational systems, e.g., black hole-neutron star systems, require physical and constraint-satisfying initial data. In this article, we present the newly developed pseudo-spectral code Elliptica, an infrastructure for construction of initial data for various binary and single gravitational systems of all kinds. The elliptic equations under consideration are solved on a single spatial hypersurface of the spacetime manifold. Using coordinate maps, the hypersurface is covered by patches whose boundaries can adapt to the surface of the compact objects. To solve elliptic equations with arbitrary boundary condition, Elliptica deploys a Schur complement domain decomposition method with a direct solver. In this version, we use cubed sphere coordinate maps and the fields are expanded using Chebyshev polynomials of the first kind. Here, we explain the building blocks of Elliptica and the initial data construction algorithm for a black hole-neutron star binary system. We perform convergence tests and evolve the data to validate our results. Within our framework, the neutron star can reach spin values close to breakup with arbitrary direction, while the black hole can have arbitrary spin with dimensionless spin magnitude $\sim 0.8$., Comment: PRD version

We study the problem of extracting a small subset of representative items from a large data stream. In many data mining and machine learning applications such as social network analysis and recommender systems, this problem can be formulated as maximizing a monotone submodular function subject to a cardinality constraint $k$. In this work, we consider the setting where data items in the stream belong to one of several disjoint groups and investigate the optimization problem with an additional \emph{fairness} constraint that limits selection to a given number of items from each group. We then propose efficient algorithms for the fairness-aware variant of the streaming submodular maximization problem. In particular, we first give a $ (\frac{1}{2}-\varepsilon) $-approximation algorithm that requires $ O(\frac{1}{\varepsilon} \log \frac{k}{\varepsilon}) $ passes over the stream for any constant $ \varepsilon>0 $. Moreover, we give a single-pass streaming algorithm that has the same approximation ratio of $(\frac{1}{2}-\varepsilon)$ when unlimited buffer sizes and post-processing time are permitted, and discuss how to adapt it to more practical settings where the buffer sizes are bounded. Finally, we demonstrate the efficiency and effectiveness of our proposed algorithms on two real-world applications, namely \emph{maximum coverage on large graphs} and \emph{personalized recommendation}., Comment: 11 pages, 8 figures, to appear in the Web conference 2021 (WWW '21)

In 2020, countries affected by the COVID-19 pandemic implemented various non-pharmaceutical interventions to contrast the spread of the virus and its impact on their healthcare systems and economies. Using Italian data at different geographic scales, we investigate the relationship between human mobility, which subsumes many facets of the population's response to the changing situation, and the spread of COVID-19. Leveraging mobile phone data from February through September 2020, we find a striking relationship between the decrease in mobility flows and the net reproduction number. We find that the time needed to switch off mobility and bring the net reproduction number below the critical threshold of 1 is about one week. Moreover, we observe a strong relationship between the number of days spent above such threshold before the lockdown-induced drop in mobility flows and the total number of infections per 100k inhabitants. Estimating the statistical effect of mobility flows on the net reproduction number over time, we document a 2-week lag positive association, strong in March and April, and weaker but still significant in June. Our study demonstrates the value of big mobility data to monitor the epidemic and inform control interventions during its unfolding.

Understanding collective mobility patterns is crucial to plan the restart of production and economic activities, which are currently put in stand-by to fight the diffusion of the epidemics. In this report, we use mobile phone data to infer the movements of people between Italian provinces and municipalities, and we analyze the incoming, outcoming and internal mobility flows before and during the national lockdown (March 9th, 2020) and after the closure of non-necessary productive and economic activities (March 23th, 2020). The population flow across provinces and municipalities enable for the modelling of a risk index tailored for the mobility of each municipality or province. Such an index would be a useful indicator to drive counter-measures in reaction to a sudden reactivation of the epidemics. Mobile phone data, even when aggregated to preserve the privacy of individuals, are a useful data source to track the evolution in time of human mobility, hence allowing for monitoring the effectiveness of control measures such as physical distancing. We address the following analytical questions: How does the mobility structure of a territory change? Do incoming and outcoming flows become more predictable during the lockdown, and what are the differences between weekdays and weekends? Can we detect proper local job markets based on human mobility flows, to eventually shape the borders of a local outbreak?

We continue our study of the binary neutron star parameter space by investigating the effect of the spin orientation on the dynamics, gravitational wave emission, and mass ejection during the binary neutron star coalescence. We simulate seven different configurations using multiple resolutions to allow a reasonable error assessment. Due to the particular choice of the setups, five configurations show precession effects, from which two show a precession ("wobbling") of the orbital plane, while three show a "bobbing" motion, i.e., the orbital angular momentum does not precess, while the orbital plane moves along the orbital angular momentum axis. Considering the ejection of mass, we find that precessing systems can have an anisotropic mass ejection, which could lead to a final remnant kick of $\sim 40 \rm km/s$ for the studied systems. Furthermore, for the chosen configurations, antialigned spins lead to larger mass ejecta than aligned spins, so that brighter electromagnetic counterparts could be expected for these configurations. Finally, we compare our simulations with the precessing, tidal waveform approximant IMRPhenomPv2_NRTidalv2 and find good agreement between the approximant and our numerical relativity waveforms with phase differences below 1.2 rad accumulated over the last $\sim$ 16 gravitational wave cycles., Comment: 18 pages, 16 figures

Neutrino-matter interactions play a key role in binary neutron star mergers. Thermodynamics conditions at the surfaces where neutrinos decouple from matter influence neutrino spectra, ultimately affecting the evolution of the remnant and the properties of the ejecta. In this work, we post-process results of general relativistic merger simulations employing microphysical equations of state and approximate neutrino transport to investigate the thermodynamics conditions at which weak and thermal equilibrium freezes out (equilibrium surfaces), as well as conditions at which the transition between diffusion and free-streaming regime occurs (diffusion surfaces). We find that the rest mass density and the neutrino energy are the most relevant quantities in determining the location of the decoupling surfaces. For mean energy neutrinos ($\langle E_{{\nu}_e} \rangle \approx 9~{\rm MeV}$, $\langle E_{{\bar{\nu}}_e} \rangle \approx 15~{\rm MeV}$, $\langle E_{{\nu}_{\mu,\tau}} \rangle \approx 25~{\rm MeV}$), diffusion surfaces are located around $10^{11}{\rm g~cm^{-3}}$ for all neutrino species, while equilibrium surfaces for heavy flavor neutrinos are significantly deeper (several $10^{12}{\rm g~cm^{-3}}$) than the ones of $\bar{\nu}_e$ and $\nu_e$ ($\gtrsim 10^{11}{\rm g~cm^{-3}}$). The resulting decoupling temperatures are in good agreement with the average neutrino energies ($\langle E_{\nu} \rangle \sim 3.15~T$), with the softer equation of state characterized by systematically larger decoupling temperatures ($\Delta T \lesssim 1~{\rm MeV}$). Neutrinos streaming at infinity with different energies come from very different regions of the remnant. The presence of a massive NS or of a BH in the remnant influences the neutrino thermalization process., Comment: 20 pages, 13 figures