Adversarially Robust Learning via Entropic Regularization

In this paper we propose a new family of algorithms, ATENT, for training adversarially robust deep neural networks. We formulate a new loss function that is equipped with an additional entropic regularization. Our loss function considers the contribution of adversarial samples that are drawn from a specially designed distribution in the data space that assigns high probability to points with high loss and in the immediate neighborhood of training samples. Our proposed algorithms optimize this loss to seek adversarially robust valleys of the loss landscape. Our approach achieves competitive (or better) performance in terms of robust classification accuracy as compared to several state-of-the-art robust learning approaches on benchmark datasets such as MNIST and CIFAR-10.

Packet loss rate monitoring model of IOT based on differential evolution algorithm

In order to overcome the problems of high energy consumption and low execution efficiency of traditional Internet of things (IOT) packet loss rate monitoring model, a new packet loss rate monitoring model based on differential evolution algorithm is proposed. The similarity between each data point in the data space of the Internet of things is set as the data gravity. On the basis of the data gravity, combined with the law of gravity in the data space, the gravity of different data is calculated. At the same time, the size of the data gravity is compared, and the data are classified. Through the classification results, the packet loss rate monitoring model of the Internet of things is established. Differential evolution algorithm is used to solve the model to obtain the best monitoring scheme to ensure the security of network data transmission. The experimental results show that the proposed model can effectively reduce the data acquisition overhead and energy consumption, and improve the execution efficiency of the model. The maximum monitoring efficiency is 99.74%.

An evaluation of the data space dimension in phase retrieval: results in Fresnel zone

<div>In this paper, we address the problem of computing the dimension of data space in phase retrieval problem. Starting from the quadratic formulation of the phase retrieval, the analysis is performed in two steps. First, we exploit the lifting technique to obtain a linear representation of the data. Later, we evaluate the dimension of data space by computing analytically the number of relevant singular values of the linear operator that represents the data. The study is done with reference to a 2D scalar geometry consisting of an electric current strip whose square amplitude of the electric radiated field is observed on a two-dimensional extended domain in Fresnel zone.</div>

Towards a Common Agricultural Data Space in the European Union: A Sustainable Development Perspective

In February 2020 the European Commission announced a new strategy for data in which an innovative proposal to create a single European data space composed of many sectoral common data spaces, including the agriculture sector, was presented. It is expected that the common agricultural data space will provide support for delivering a smart, innovative and sustainable agri-food system from farm to fork. Based on the analysis of framework conditions for pooling and sharing agricultural data in the EU and the Commission’s initiatives in this area, this article aims to discuss how and to what extent the common data space in agriculture could contribute to environmental, economic and social sustainability in the EU. It was concluded that the achievement of sustainability goals with the help of the planned common data space remains challenging, particularly in the context of rapid, but uneven pace of digital transformation in the agri-food sector in the EU. Overcoming legal, technical and other barriers to data sharing in the EU will not remove the fundamental problems of limited representativeness of current agricultural data assets in the EU. The design of the common data space in agriculture as well as the rules for data access and use should therefore be carefully considered. Also, specific and datarelated intervention measures, e.g. under the CAP, would be needed both to decrease the problem of a fragmented farm data landscape and to respond to the growing needs to collect and share private farm data that are highly relevant to achieving broader social goals and sustainability.