With the rise of social networks and the introduction of data protection laws, companies are training machine learning models using data generated locally by their users or customers in various types of devices. The data may include sensitive information such as family information, medical records, personal habits, or financial records that, if leaked, can generate problems. For this reason, this paper aims to introduce a protocol for training Multi-Layer Perceptron (MLP) neural networks via combining federated learning and homomorphic encryption, where the data are distributed in multiple clients, and the data privacy is preserved. This proposal was validated by running several simulations using a dataset for a multi-class classification problem, different MLP neural network architectures, and different numbers of participating clients. The results are shown for several metrics in the local and federated settings, and a comparative analysis is carried out. Additionally, the privacy guarantees of the proposal are formally analyzed under a set of defined assumptions, and the added value of the proposed protocol is identified compared with previous works in the same area of knowledge.
Abstract-This paper presents a system of recommendations for an enterprise content manager (ECM) based on ontological models. In many occasions the results of a search are not accurate enough, so the user of the ECM system must check them and discard those not related to the search. In order to make recommendations, a proposal where it is necessary to review the instances of the ontological model is presented to manage the alias and ambiguities. Comparisons are made between the results obtained from the traditional search model and the recommendations suggested by the model proposed in this work.
A one-session multicast network, on which a coding scheme with Network Coding is defined, was implemented with a maximum common flow of r-packets arriving simultaneously at |T | sink nodes. Determining how to order the r-packets that emerge from the source node s through their output n-links, constitutes a combinatorial problem. In this work, the set of all the possible output configurations is constructed, where each configuration is a vector of packets tags of length equal to n. Each tag has a length equal to r. Through a combinatorial algorithm on the set of possible output configurations, a path is carried out on the graph representing the one-session multicast network. The path is based on a topological ordering of the multicast graph that allowed us finding all possible ways to order the output of the r-packets from s to the sink nodes in T. An ordering configuration based on Network Coding is valid, if the coding of packets is achieved through a linear combination in the coding nodes and the decoding of packets in the sink nodes. This validation verifies, then, a one-session multicast solution. The proposal of this work is independent of the network topology, the maximum flow value, and the size of the packets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.