Towards developmental Machine Learning
Marco Gori, University of Siena
By and large, most studies of machine learning and pattern recognition are rooted in the framework of statistics. This is primarily due to the way machine learning is traditionally posed, namely by a problem of extraction of regularities from a sample
of a probability distribution. This course promotes a truly different way of interpreting the learning of that relies on system dynamics. We promote a view of learning as the outcome of laws of nature that govern the interactions
of intelligent agents with their own environment.This leads to an in-depth interpretation of causality along with the definition of the principles and the methods for learning to store events without their long-term forgetting
that characterize state of the art technologies in recurrent neural networks. Finally, we reinforce the underlying principle that the acquisition of cognitive skills by learning obeys information-based laws based on variational
principles, which hold regardless of biology.
Opinions and conflict in social networks: models, computational problems, and algorithms
Aristides Gionis, KTH Royal Institute of Technology
Online social networks are important venues of public discourse today, hosting the opinions of hundreds of millions of individuals. Social networks are often credited for providing a technological means to break information barriers and promote diversity
and democracy. In practice, however, the opposite effect is often observed: users tend to favor content that agrees with their existing world-view, get less exposure to conflicting viewpoints, and eventually create "echo chambers"
and increased polarization. Arguably, without any kind of moderation, current social-networking platforms gravitate towards a state in which net-citizens are constantly reinforcing their existing opinions. In this course we present
a systematic review of polarization as manifested online, and in particular in online social networks. We start by defining the concept of polarization and reviewing algorithmic methods for detecting, quantifying, and mitigating
polarization. Subsequently, we provide an overview of the theory of signed networks, where edges are labeled by a sign, positive or negative. In a social network, where edges might represent interactions between users, the sign
may determine whether an exchange was friendly or hostile. This simple modification to the standard graph model gives rise to interesting problem formulations and algorithmic techniques in the context of studying polarization in
social networks. Finally, we will discuss models proposed in the literature to explain how individuals form opinions in social networks. We will present the most important opinion-formation models and will discuss some of the computational
challenges that have arisen recently.
From Cloud to Serverless through microelements
Recent technological advances have disrupted the current centralized cloud computing model, moving cloud resources close to users. Microservice approach allows to instantiate a new paradigm that’s driven by the significant increase in resource capacity/capability
at the network fog/edge, along with support for data transfer protocols that enable such resources to interact more seamlessly with datacenter-based services. One of the near future challenges is represented by the management of
even more high distributed systems along with the need to federate those environments. The automatic deployment of microservices is becoming a must, importantly they have to be composed and interconnected over IoT, Fog, edge and
cloud infrastructures, adopting the Serverless paradigm, and also thinking to Security from the beginning. Various stakeholders (Cloud providers, Edge providers, Fog provider, Security provider, Application providers, IoT, DevOPs,
and so on) can contribute to the provisioning of new Serverless and FaaS applications in Federated Environments. Osmotic Computing relies on microelements, an extension of microservices that Serverless can benefit of, for smoothly
managing challenging services of the future.