Design and implementation of a Reinforcement Learning framework for iOS devices

Reinforcement Learning is an increasingly popular area of Artificial Intelligence. The applications of this learning paradigm are many, but its application in mobile computing is in its infancy. This study aims to provide an overview of current Reinforcement Learning applications on mobile devices, as well as to introduce a new framework for iOS devices: Swift-RL Lib. This new Swift package allows developers to easily support and integrate two of the most common RL algorithms, Q-Learning and Deep Q-Network, in a fully customizable environment. All processes are performed on the device, without any need for remote computation. The framework was tested in different settings and evaluated through several use cases. Through an in-depth performance analysis, we show that the platform provides effective and efficient support for Reinforcement Learning for mobile applications.

No(i)rthern Ireland: Crime fiction and the northern-irish scene. Proposed translation into italian of two short stories from "belfast noir"

The present thesis aims at proving the importance of cultural and literary contexts in the practice of translation: I shall show that, in the case of Northern Irish crime fiction, knowledge of both Northern Irish history and culture as well as of the genre of crime fiction are essential prerequisites for the production of a “responsible” translation. I will therefore offer a brief overview of the history of crime and detective fiction and its main subgenres; some of the most important authors and works will be presented as well, in an analysis that goes from the early years of the genre to the second half of the 20th century. I will then move the focus to Northern Ireland, its culture and its history, and particular attention will be paid to fiction writing in Ireland and Northern Ireland, with a focus on the peculiar phenomenon of “Troubles Trash”. I will tackle the topic of Northern Irish literature and present the contemporary scene of Northern Irish crime fiction; the volume from which the texts for the translation have been taken will be presented, namely Belfast Noir. Subsequently the focus will move on the theoretical framework within which the translations were produced: I will present a literary review of the most significative developments in Translation Studies, with particular attention to the “cultural turn” that has characterised this subject since the 1960s. I will then highlight the phenomenon of “realia” in translation and analyse the approaches of different scholars to the translation of culture-bound references. The final part represents the culmination and practical application of all that was presented in the previous sections: I will discuss the translation of culture-bound references according to the strategies presented in Chapter 4, referring to the proposed translations of two stories. Such analysis aims to show that not only expert linguistic knowledge, but also cultural awareness and a wide literary background are needed in order to make conscious choices in translation.

Machine learning "as a service" for high energy physics (MLaaS4HEP): evolution of a framework for ML-based physics analyses

The scientific success of the LHC experiments at CERN highly depends on the availability of computing resources which efficiently store, process, and analyse the amount of data collected every year. This is ensured by the Worldwide LHC Computing Grid infrastructure that connect computing centres distributed all over the world with high performance network. LHC has an ambitious experimental program for the coming years, which includes large investments and improvements both for the hardware of the detectors and for the software and computing systems, in order to deal with the huge increase in the event rate expected from the High Luminosity LHC (HL-LHC) phase and consequently with the huge amount of data that will be produced. Since few years the role of Artificial Intelligence has become relevant in the High Energy Physics (HEP) world. Machine Learning (ML) and Deep Learning algorithms have been successfully used in many areas of HEP, like online and offline reconstruction programs, detector simulation, object reconstruction, identification, Monte Carlo generation, and surely they will be crucial in the HL-LHC phase. This thesis aims at contributing to a CMS R&D project, regarding a ML "as a Service" solution for HEP needs (MLaaS4HEP). It consists in a data-service able to perform an entire ML pipeline (in terms of reading data, processing data, training ML models, serving predictions) in a completely model-agnostic fashion, directly using ROOT files of arbitrary size from local or distributed data sources. This framework has been updated adding new features in the data preprocessing phase, allowing more flexibility to the user. Since the MLaaS4HEP framework is experiment agnostic, the ATLAS Higgs Boson ML challenge has been chosen as physics use case, with the aim to test MLaaS4HEP and the contribution done with this work.

Learning from experience: Analysis of the civil litigation regarding building structures on the court of bologna in the years 2011/2012

The present work is aimed to the study and the analysis of the defects detected in the civil structure and that are object of civil litigation in order to create an instruments capable of helping the different actor involved in the building process. It is divided in three main sections. The first part is focused on the collection of the data related to the civil proceeding of the 2012 and the development of in depth analysis of the main aspects regarding the defects on existing buildings. The research center “Osservatorio Claudio Ceccoli” developed a system for the collection of the information coming from the civil proceedings of the Court of Bologna. Statistical analysis are been performed and the results are been shown and discussed in the first chapters.The second part analyzes the main issues emerged during the study of the real cases, related to the activities of the technical consultant. The idea is to create documents, called “focus”, addressed to clarify and codify specific problems in order to develop guidelines that help the technician editing of the technical advice.The third part is centered on the estimation of the methods used for the collection of data. The first results show that these are not efficient. The critical analysis of the database, the result and the experience and throughout, allowed the implementation of the collection system for the data.

Dynamic Human Robot Interaction Framework Using Deep Learning and Robot Operating System (ROS): a practical approach

Trying to explain to a robot what to do is a difficult undertaking, and only specific types of people have been able to do so far, such as programmers or operators who have learned how to use controllers to communicate with a robot. My internship's goal was to create and develop a framework that would make that easier. The system uses deep learning techniques to recognize a set of hand gestures, both static and dynamic. Then, based on the gesture, it sends a command to a robot. To be as generic as feasible, the communication is implemented using Robot Operating System (ROS). Furthermore, users can add new recognizable gestures and link them to new robot actions; a finite state automaton enforces the users' input verification and correct action sequence. Finally, the users can create and utilize a macro to describe a sequence of actions performable by a robot.

Meta-learning for informed model searches in Automated Machine Learning

Day by day, machine learning is changing our lives in ways we could not have imagined just 5 years ago. ML expertise is more and more requested and needed, though just a limited number of ML engineers are available on the job market, and their knowledge is always limited by an inherent characteristic of theirs: they are humans. This thesis explores the possibilities offered by meta-learning, a new field in ML that takes learning a level higher: models are trained on other models' training data, starting from features of the dataset they were trained on, inference times, obtained performances, to try to understand the relationship between a good model and the way it was obtained. The so-called metamodel was trained on data collected by OpenML, the largest ML metadata platform that's publicly available today. Datasets were analyzed to obtain meta-features that describe them, which were then tied to model performances in a regression task. The obtained metamodel predicts the expected performances of a given model type (e.g., a random forest) on a given ML task (e.g., classification on the UCI census dataset). This research was then integrated into a custom-made AutoML framework, to show how meta-learning is not an end in itself, but it can be used to further progress our ML research. Encoding ML engineering expertise in a model allows better, faster, and more impactful ML applications across the whole world, while reducing the cost that is inevitably tied to human engineers.