1,721,238 research outputs found
Entropic Regularization in Hierarchical Games
No full textWe study hierarchical games where the second stage consists of a finite noncoopera- tive game. To ensure that the lower level problem admits solutions, its mixed exten- sion is considered. By using the Shannon entropy, a regularization scheme for the two-stage game is introduced and some properties are presented, as the asymptotic subgame perfectness
Lazy Network: A Word Embedding-Based Temporal Financial Network to Avoid Economic Shocks in Asset Pricing Models
Get PDFPublic companies in the US stock market must annually report their activities and financial performances to the SEC by filing the
so-called 10-K form. Recent studies have demonstrated that changes in the textual content of the corporate annual filing (10-K)
can convey strong signals of companies’ future returns. In this study, we combine natural language processing techniques and
network science to introduce a novel 10-K-based network, named Lazy Network, that leverages year-on-year changes in
companies’ 10-Ks detected using a neural network embedding model. (e Lazy Network aims to capture textual changes derived
from financial or economic changes on the equity market. Leveraging the Lazy Network, we present a novel investment strategy
that attempts to select the least disrupted and stable companies by capturing the peripheries of the Lazy Network. We show that
this strategy earns statistically significant risk-adjusted excess returns. Specifically, the proposed portfolios yield up to 95 basis
points in monthly five-factor alphas (over 12% annually), outperforming similar strategies in the literature
Leveraging Smart City Projects for Benefitting Citizens: The Role of ICTs
No full textThe scientific literature converges in indicating better life conditions for citizens as the smart cityâ s main goal. To achieve this goal, cities leverage different technologies and especially ICT to modify urban infrastructures, public and private services and governance activities. However, smart programs often target the use and experimentation of innovative technologies, whilst citizens are considered as the passive addressees of technological programs. To verify whether smart projects really pursue citizensâ well-being, an extensive empirical survey has been conducted. The research investigates 366 European smart city projects and extracts 42 ICT-enabled projects explicitly focusing on citizens. The analysis sheds light on the complex goal of citizensâ well-being improvement in smart cities and on the most promising ICT solutions to impact urban life conditions. A special focus regards the use of IoT in smart projects addressing the citizensâ well-being
CliSAT: A new exact algorithm for hard maximum clique problems
No full textGiven a graph, the maximum clique problem (MCP) asks for determining a complete subgraph with the largest possible number of vertices. We propose a new exact algorithm, called CliSAT, to solve the MCP to proven optimality. This problem is of fundamental importance in graph theory and combinatorial optimization due to its practical relevance for a wide range of applications. The newly developed exact approach is a combinatorial branch-and-bound algorithm that exploits the state-of-the-art branching scheme enhanced by two new bounding techniques with the goal of reducing the branching tree. The first one is based on graph colouring procedures and partial maximum satisfiability problems arising in the branching scheme. The second one is a filtering phase based on constraint programming and domain propagation techniques. CliSAT is designed for structured MCP instances which are computationally difficult to solve since they are dense and contain many interconnected large cliques. Extensive experiments on hard benchmark instances, as well as new hard instances arising from different applications, show that CliSAT outperforms the state-of-the-art MCP algorithms, in some cases by several orders of magnitude.This publication is part of the R&D project “Cognitive Personal Assistance for Social Environments (ACOGES)”, reference PID2020-113096RB-I00, funded by MCIN/AEI/10.13039/501100011033.Peer reviewe
The Perturbed Median Principle for Integral Inequalities with Applications
Get PDFIn this paper a perturbed version of the Median Principle introduced
by the author in 'The median principle for inequalities and applications' is developed. Applications for various Riemann-
Stieltjes integral and Lebesgue integral inequalities are also provided
Informetric models for citation frequency data: an empirical investigation
No full textThis paper tries to answer to the question “Which is the best model for representing the citation frequency curve?”. We consider 131 practical cases of physicists who were applicants for a full professorship in the specific area of Condensed Matter Physics, and we estimate a set of four possible different types of distribution for size-frequency data. The most remarkable result is that the well-known Lotkaian model is not the best fitting among all we have considered. From our data we conclude that the geometric distribution can provide a valid alternative to more traditional models
Neural network embedding: representation learning and latent knowledge extraction for data mining applications
No full textBig data represents what machine learning models need to learn concepts and tasks, providing enough generalization margins. Moreover, it can also be useful for data mining applications when the goal is to extract latent and valuable knowledge. The nature of data we can collect every day is unstructured: no predefined schema ispresent. Some examples are documents, messages, interactions in social media platforms and in e-commerce web sites. Indeed, the unstructured nature of these data, inevitably, adds more challenges in the previously reported cases. Machine learning models need input data in real-valued vectors (feature or design matrix) for supervised and unsupervised learning. The same issue comes back again when we are interested in finding quantitative information (e.g., the similarity between two or more documents or measure any statistics). Another interesting case is the one related to network science techniques to analyze data having (or looking for) a graph structure. Several systems can be efficiently described as nodes that interact with each other (e.g., social networks, recommendation systems, interaction among proteins, financial market). In this case, we cannot apply machine learning techniques directly because of the lack of a vector representation.The advances in the neural networks field, enable to learn feature vectors directly from the input data distribution. This task can be seen as a pre-processing step in a modern machine learning project and involve machine learning itself. This automatic feature extraction is also known as “Representation learning”. We can summarize itas learning a vector representation of input data in a supervised or unsupervised way.
ivAbstractWe can learn these representations or embeddings for different kinds of data: from words, entire documents, nodes or edges in a graph, images, and signals. Moreover, embeddings encode data preserving and providing more information, for example, semantic similarity for words that will be close in their vector representation.The choice of how to achieve this preliminary task influences all of the next stages in a typical data-mining pipeline. The challenge consists into finding a way to preserve much information as possible when looking for a structure. Secondary, with the promising results achieved recently, it is interesting to exploit these resulting vector spaces also to make knowledge extraction. In light of the current challenges and advances in the field of Representation Learn-ing with unstructured data, my research activity has been focused on this topic. In particular, in this thesis are reported the achieved results in two main directions:•Using representation learning techniques based on neural networks to extract latent knowledge from documents: If neural networks can capture fundamental aspects among data by learning a different representation in the out-put layer and considering that this representation makes the classification or clustering easier, can we exploit these techniques to extract new knowledge? Part of my research tries to answer to this question. In particular, two use cases are reported: the first analyzes scientific documents from the public repositoryScopus combining word embeddings and Human mobility metrics. The second one regards neural network embeddings to extract knowledge from financial reports of thousands of American companies in the stock market.•Overcome current limits of representation learning on graphs: Machine learning models can benefit from input data derived from graph structures. However, to apply most of the available models it is necessary to get a vector form for nodes and edges. The most promising way is related to the neural
Abstractvnetwork embedding and the state of the art is represented by the Node2vecalgorithm. However, there are still two open problems in this field: scalability (learning a representation in large-scale graphs) and the lack of support of dynamic contexts: if a new node joins the network it is necessary to compute again the representation of the entire graph. Part of my doctorate tries to address these two problems. A first contribution is represented by an actor-basedversion of Node2Vec that overcomes scalability issues by distributing the bot-tlenecks with agents that organize themself with different behaviors to achievethe embedding in large-scale graphs. A second contribution is related to the de-velopment of a novel algorithm for incremental feature learning over graphs.The algorithm exploits properties of scale-free graphs to encode new nodeswithout recurring to a re-train of the model over all the nodes. It computes alight embedding over 20% of nodes with the highest degree, and then it per-forms a supervised alignment by solving the orthogonal Procrustes problem
Engineering the modulo network simplex heuristic for the periodic timetabling problem
No full textThe Periodic Event Scheduling Problem (PESP), in which events have to be scheduled repeatedly over a given period, is a complex and well-known discrete problem with numerous real-world applications. One of them is to find periodic timetables which is economically important, but difficult to handle mathematically, since even finding a feasible solution to this problem is known to be NP-hard. On the other hand, there are recent achievements like the computation of the timetable of the Dutch railway system that impressively demonstrate the applicability and practicability of the mathematical model. In this paper we propose different approaches to improve the modulo network simplex algorithm [8], which is a powerful heuristic for the PESP problem, by exploiting improved search methods in the modulo simplex tableau and larger classes of cuts to escape from the many local optima. Numerical experiments on railway instances show that our algorithms are able to handle problems of the size of the German intercity railway network
- …
