arXiv subjects

Bruno Lepri

Publications and source records attributed to Bruno Lepri.

At least 109 records · Page 6Linked to original sources

arXiv2022

A Framework for Verifiable and Auditable Federated Anomaly Detection

Federated Leaning is an emerging approach to manage cooperation between a group of agents for the solution of Machine Learning tasks, with the goal of improving each agent's performance without disclosing any data. In this paper we present a novel algorithmic architecture that tackle this problem in the particular case of Anomaly Detection (or classification or rare events), a setting where typical applications often comprise data with sensible information, but where the scarcity of anomalous examples encourages collaboration. We show how Random Forests can be used as a tool for the development of accurate classifiers with an effective insight-sharing mechanism that does not break the data integrity. Moreover, we explain how the new architecture can be readily integrated in a blockchain infrastructure to ensure the verifiable and auditable execution of the algorithm. Furthermore, we discuss how this work may set the basis for a more general approach for the design of federated ensemble-learning methods beyond the specific task and architecture discussed in this paper.

cs.LG↗

arXiv2022

Reprogramming FairGANs with Variational Auto-Encoders: A New Transfer Learning Model

Fairness-aware GANs (FairGANs) exploit the mechanisms of Generative Adversarial Networks (GANs) to impose fairness on the generated data, freeing them from both disparate impact and disparate treatment. Given the model's advantages and performance, we introduce a novel learning framework to transfer a pre-trained FairGAN to other tasks. This reprogramming process has the goal of maintaining the FairGAN's main targets of data utility, classification utility, and data fairness, while widening its applicability and ease of use. In this paper we present the technical extensions required to adapt the original architecture to this new framework (and in particular the use of Variational Auto-Encoders), and discuss the benefits, trade-offs, and limitations of the new model.

cs.LG↗

arXiv2022

Trajectory Test-Train Overlap in Next-Location Prediction Datasets

Next-location prediction, consisting of forecasting a user's location given their historical trajectories, has important implications in several fields, such as urban planning, geo-marketing, and disease spreading. Several predictors have been proposed in the last few years to address it, including last-generation ones based on deep learning. This paper tests the generalization capability of these predictors on public mobility datasets, stratifying the datasets by whether the trajectories in the test set also appear fully or partially in the training set. We consistently discover a severe problem of trajectory overlapping in all analyzed datasets, highlighting that predictors memorize trajectories while having limited generalization capacities. We thus propose a methodology to rerank the outputs of the next-location predictors based on spatial mobility patterns. With these techniques, we significantly improve the predictors' generalization capability, with a relative improvement on the accuracy up to 96.15% on the trajectories that cannot be memorized (i.e., low overlap with the training set).

cs.AI↗

arXiv2022

ISF-GAN: An Implicit Style Function for High-Resolution Image-to-Image Translation

Recently, there has been an increasing interest in image editing methods that employ pre-trained unconditional image generators (e.g., StyleGAN). However, applying these methods to translate images to multiple visual domains remains challenging. Existing works do not often preserve the domain-invariant part of the image (e.g., the identity in human face translations), they do not usually handle multiple domains, or do not allow for multi-modal translations. This work proposes an implicit style function (ISF) to straightforwardly achieve multi-modal and multi-domain image-to-image translation from pre-trained unconditional generators. The ISF manipulates the semantics of an input latent code to make the image generated from it lying in the desired visual domain. Our results in human face and animal manipulations show significantly improved results over the baselines. Our model enables cost-effective multi-modal unsupervised image-to-image translations at high resolution using pre-trained unconditional GANs. The code and data are available at: \url{https://github.com/yhlleo/stylegan-mmuit}.

cs.CV↗

arXiv2021

Efficient Training of Visual Transformers with Small Datasets

Visual Transformers (VTs) are emerging as an architectural paradigm alternative to Convolutional networks (CNNs). Differently from CNNs, VTs can capture global relations between image elements and they potentially have a larger representation capacity. However, the lack of the typical convolutional inductive bias makes these models more data-hungry than common CNNs. In fact, some local properties of the visual domain which are embedded in the CNN architectural design, in VTs should be learned from samples. In this paper, we empirically analyse different VTs, comparing their robustness in a small training-set regime, and we show that, despite having a comparable accuracy when trained on ImageNet, their performance on smaller datasets can be largely different. Moreover, we propose a self-supervised task which can extract additional information from images with only a negligible computational overhead. This task encourages the VTs to learn spatial relations within an image and makes the VT training much more robust when training data are scarce. Our task is used jointly with the standard (supervised) training and it does not depend on specific architectural choices, thus it can be easily plugged in the existing VTs. Using an extensive evaluation with different VTs and datasets, we show that our method can improve (sometimes dramatically) the final accuracy of the VTs. Our code is available at: https://github.com/yhlleo/VTs-Drloc.

cs.CV↗

arXiv2021

An Efficient Procedure for Mining Egocentric Temporal Motifs

Temporal graphs are structures which model relational data between entities that change over time. Due to the complex structure of data, mining statistically significant temporal subgraphs, also known as temporal motifs, is a challenging task. In this work, we present an efficient technique for extracting temporal motifs in temporal networks. Our method is based on the novel notion of egocentric temporal neighborhoods, namely multi-layer structures centered on an ego node. Each temporal layer of the structure consists of the first-order neighborhood of the ego node, and corresponding nodes in sequential layers are connected by an edge. The strength of this approach lies in the possibility of encoding these structures into a unique bit vector, thus bypassing the problem of graph isomorphism in searching for temporal motifs. This allows our algorithm to mine substantially larger motifs with respect to alternative approaches. Furthermore, by bringing the focus on the temporal dynamics of the interactions of a specific node, our model allows to mine temporal motifs which are visibly interpretable. Experiments on a number of complex networks of social interactions confirm the advantage of the proposed approach over alternative non-egocentric solutions. The egocentric procedure is indeed more efficient in revealing similarities and discrepancies among different social environments, independently of the different technologies used to collect data, which instead affect standard non-egocentric measures.

cs.SI↗

arXiv2021

Click to Move: Controlling Video Generation with Sparse Motion

This paper introduces Click to Move (C2M), a novel framework for video generation where the user can control the motion of the synthesized video through mouse clicks specifying simple object trajectories of the key objects in the scene. Our model receives as input an initial frame, its corresponding segmentation map and the sparse motion vectors encoding the input provided by the user. It outputs a plausible video sequence starting from the given frame and with a motion that is consistent with user input. Notably, our proposed deep architecture incorporates a Graph Convolution Network (GCN) modelling the movements of all the objects in the scene in a holistic manner and effectively combining the sparse user motion information and image features. Experimental results show that C2M outperforms existing methods on two publicly available datasets, thus demonstrating the effectiveness of our GCN framework at modelling object interactions. The source code is publicly available at https://github.com/PierfrancescoArdino/C2M.

cs.CV↗

arXiv2021

Living in a pandemic: adaptation of individual mobility and social activity in the US

The non-pharmaceutical interventions (NPIs), aimed at reducing the diffusion of the COVID-19 pandemic, has dramatically influenced our behaviour in everyday life. In this work, we study how individuals adapted their daily movements and person-to-person contact patterns over time in response to the COVID-19 pandemic and the NPIs. We leverage longitudinal GPS mobility data of hundreds of thousands of anonymous individuals in four US states and empirically show the dramatic disruption in people's life. We find that local interventions did not just impact the number of visits to different venues but also how people experience them. Individuals spend less time in venues, preferring simpler and more predictable routines and reducing person-to-person contact activities. Moreover, we show that the stringency of interventions alone does explain the number and duration of visits to venues: individual patterns of visits seem to be influenced by the local severity of the pandemic and a risk adaptation factor, which increases the people's mobility regardless of the stringency of interventions.

cs.CY↗

arXiv2021

Optimizing city-scale traffic through modeling observations of vehicle movements

The capability of traffic-information systems to sense the movement of millions of users and offer trip plans through mobile phones has enabled a new way of optimizing city traffic dynamics, turning transportation big data into insights and actions in a closed-loop and evaluating this approach in the real world. Existing research has applied dynamic Bayesian networks and deep neural networks to make traffic predictions from floating car data, utilized dynamic programming and simulation approaches to identify how people normally travel with dynamic traffic assignment for policy research, and introduced Markov decision processes and reinforcement learning to optimally control traffic signals. However, none of these works utilized floating car data to suggest departure times and route choices in order to optimize city traffic dynamics. In this paper, we present a study showing that floating car data can lead to lower average trip time, higher on-time arrival ratio, and higher Charypar-Nagel score compared with how people normally travel. The study is based on optimizing a partially observable discrete-time decision process and is evaluated in one synthesized scenario, one partly synthesized scenario, and three real-world scenarios. This study points to the potential of a "living lab" approach where we learn, predict, and optimize behaviors in the real world.

cs.SI↗

arXiv2021

A Survey on Deep Learning for Human Mobility

The study of human mobility is crucial due to its impact on several aspects of our society, such as disease spreading, urban planning, well-being, pollution, and more. The proliferation of digital mobility data, such as phone records, GPS traces, and social media posts, combined with the predictive power of artificial intelligence, triggered the application of deep learning to human mobility. Existing surveys focus on single tasks, data sources, mechanistic or traditional machine learning approaches, while a comprehensive description of deep learning solutions is missing. This survey provides a taxonomy of mobility tasks, a discussion on the challenges related to each task and how deep learning may overcome the limitations of traditional models, a description of the most relevant solutions to the mobility tasks described above and the relevant challenges for the future. Our survey is a guide to the leading deep learning solutions to next-location prediction, crowd flow prediction, trajectory generation, and flow generation. At the same time, it helps deep learning scientists and practitioners understand the fundamental concepts and the open challenges of the study of human mobility.

cs.LG↗

arXiv2021

Smoothing the Disentangled Latent Style Space for Unsupervised Image-to-Image Translation

Image-to-Image (I2I) multi-domain translation models are usually evaluated also using the quality of their semantic interpolation results. However, state-of-the-art models frequently show abrupt changes in the image appearance during interpolation, and usually perform poorly in interpolations across domains. In this paper, we propose a new training protocol based on three specific losses which help a translation network to learn a smooth and disentangled latent style space in which: 1) Both intra- and inter-domain interpolations correspond to gradual changes in the generated images and 2) The content of the source image is better preserved during the translation. Moreover, we propose a novel evaluation metric to properly measure the smoothness of latent style space of I2I translation models. The proposed method can be plugged into existing translation approaches, and our extensive experiments on different datasets show that it can significantly boost the quality of the generated images and the graduality of the interpolations.

cs.CV↗

arXiv2021

Leveraging Mobile Phone Data for Migration Flows

Statistics on migration flows are often derived from census data, which suffer from intrinsic limitations, including costs and infrequent sampling. When censuses are used, there is typically a time gap - up to a few years - between the data collection process and the computation and publication of relevant statistics. This gap is a significant drawback for the analysis of a phenomenon that is continuously and rapidly changing. Alternative data sources, such as surveys and field observations, also suffer from reliability, costs, and scale limitations. The ubiquity of mobile phones enables an accurate and efficient collection of up-to-date data related to migration. Indeed, passively collected data by the mobile network infrastructure via aggregated, pseudonymized Call Detail Records (CDRs) is of great value to understand human migrations. Through the analysis of mobile phone data, we can shed light on the mobility patterns of migrants, detect spontaneous settlements and understand the daily habits, levels of integration, and human connections of such vulnerable social groups. This Chapter discusses the importance of leveraging mobile phone data as an alternative data source to gather precious and previously unavailable insights on various aspects of migration. Also, we highlight pending challenges that would need to be addressed before we can effectively benefit from the availability of mobile phone data to help make better decisions that would ultimately improve millions of people's lives.

cs.CY↗

arXiv2021

Detecting discriminatory risk through data annotation based on Bayesian inferences

Thanks to the increasing growth of computational power and data availability, the research in machine learning has advanced with tremendous rapidity. Nowadays, the majority of automatic decision making systems are based on data. However, it is well known that machine learning systems can present problematic results if they are built on partial or incomplete data. In fact, in recent years several studies have found a convergence of issues related to the ethics and transparency of these systems in the process of data collection and how they are recorded. Although the process of rigorous data collection and analysis is fundamental in the model design, this step is still largely overlooked by the machine learning community. For this reason, we propose a method of data annotation based on Bayesian statistical inference that aims to warn about the risk of discriminatory results of a given data set. In particular, our method aims to deepen knowledge and promote awareness about the sampling practices employed to create the training set, highlighting that the probability of success or failure conditioned to a minority membership is given by the structure of the data available. We empirically test our system on three datasets commonly accessed by the machine learning community and we investigate the risk of racial discrimination.

cs.LG↗

arXiv2020

From code to market: Network of developers and correlated returns of cryptocurrencies

"Code is law" is the funding principle of cryptocurrencies. The security, transferability, availability and other properties of a crypto-asset are determined by the code through which it is created. If code is open source, as it happens for most cryptocurrencies, this principle would prevent manipulations and grant transparency to users and traders. However, this approach considers cryptocurrencies as isolated entities thus neglecting possible connections between them. Here, we show that 4% of developers contribute to the code of more than one cryptocurrency and that the market reflects these cross-asset dependencies. In particular, we reveal that the first coding event linking two cryptocurrencies through a common developer leads to the synchronisation of their returns in the following months. Our results identify a clear link between the collaborative development of cryptocurrencies and their market behaviour. More broadly, our work reveals a so-far overlooked systemic dimension for the transparency of code-based ecosystems and we anticipate it will be of interest to researchers, investors and regulators.

q-fin.ST↗

arXiv2020

Semantic-Guided Inpainting Network for Complex Urban Scenes Manipulation

Manipulating images of complex scenes to reconstruct, insert and/or remove specific object instances is a challenging task. Complex scenes contain multiple semantics and objects, which are frequently cluttered or ambiguous, thus hampering the performance of inpainting models. Conventional techniques often rely on structural information such as object contours in multi-stage approaches that generate unreliable results and boundaries. In this work, we propose a novel deep learning model to alter a complex urban scene by removing a user-specified portion of the image and coherently inserting a new object (e.g. car or pedestrian) in that scene. Inspired by recent works on image inpainting, our proposed method leverages the semantic segmentation to model the content and structure of the image, and learn the best shape and location of the object to insert. To generate reliable results, we design a new decoder block that combines the semantic segmentation and generation task to guide better the generation of new objects and scenes, which have to be semantically consistent with the image. Our experiments, conducted on two large-scale datasets of urban scenes (Cityscapes and Indian Driving), show that our proposed approach successfully address the problem of semantically-guided inpainting of complex urban scene.

cs.CV↗

arXiv2020

Temporal properties of higher-order interactions in social networks

Human social interactions in local settings can be experimentally detected by recording the physical proximity and orientation of people. Such interactions, approximating face-to-face communications, can be effectively represented as time varying social networks with links being unceasingly created and destroyed over time. Traditional analyses of temporal networks have addressed mostly pairwise interactions, where links describe dyadic connections among individuals. However, many network dynamics are hardly ascribable to pairwise settings but often comprise larger groups, which are better described by higher-order interactions. Here we investigate the higher-order organizations of temporal social networks by analyzing three publicly available datasets collected in different social settings. We find that higher-order interactions are ubiquitous and, similarly to their pairwise counterparts, characterized by heterogeneous dynamics, with bursty trains of rapidly recurring higher-order events separated by long periods of inactivity. We investigate the evolution and formation of groups by looking at the transition rates between different higher-order structures. We find that in more spontaneous social settings, group are characterized by slower formation and disaggregation, while in work settings these phenomena are more abrupt, possibly reflecting pre-organized social dynamics. Finally, we observe temporal reinforcement suggesting that the longer a group stays together the higher the probability that the same interaction pattern persist in the future. Our findings suggest the importance of considering the higher-order structure of social interactions when investigating human temporal dynamics.

physics.soc-ph↗

arXiv2020

Retrieval Guided Unsupervised Multi-domain Image-to-Image Translation

Image to image translation aims to learn a mapping that transforms an image from one visual domain to another. Recent works assume that images descriptors can be disentangled into a domain-invariant content representation and a domain-specific style representation. Thus, translation models seek to preserve the content of source images while changing the style to a target visual domain. However, synthesizing new images is extremely challenging especially in multi-domain translations, as the network has to compose content and style to generate reliable and diverse images in multiple domains. In this paper we propose the use of an image retrieval system to assist the image-to-image translation task. First, we train an image-to-image translation model to map images to multiple domains. Then, we train an image retrieval model using real and generated images to find images similar to a query one in content but in a different domain. Finally, we exploit the image retrieval system to fine-tune the image-to-image translation model and generate higher quality images. Our experiments show the effectiveness of the proposed solution and highlight the contribution of the retrieval network, which can benefit from additional unlabeled data and help image-to-image translation models in the presence of scarce data.

cs.CV↗

arXiv2020

Describe What to Change: A Text-guided Unsupervised Image-to-Image Translation Approach

Manipulating visual attributes of images through human-written text is a very challenging task. On the one hand, models have to learn the manipulation without the ground truth of the desired output. On the other hand, models have to deal with the inherent ambiguity of natural language. Previous research usually requires either the user to describe all the characteristics of the desired image or to use richly-annotated image captioning datasets. In this work, we propose a novel unsupervised approach, based on image-to-image translation, that alters the attributes of a given image through a command-like sentence such as "change the hair color to black". Contrarily to state-of-the-art approaches, our model does not require a human-annotated dataset nor a textual description of all the attributes of the desired image, but only those that have to be modified. Our proposed model disentangles the image content from the visual attributes, and it learns to modify the latter using the textual description, before generating a new image from the content and the modified attribute representation. Because text might be inherently ambiguous (blond hair may refer to different shadows of blond, e.g. golden, icy, sandy), our method generates multiple stochastic versions of the same translation. Experiments show that the proposed model achieves promising performances on two large-scale public datasets: CelebA and CUB. We believe our approach will pave the way to new avenues of research combining textual and speech commands with visual attributes.

cs.CV↗