Augmented Audiovisual Translation: the Perks and Perils of the Implementation of Artificial Intelligence in Subtitling and Dubbing

This thesis examines the state of audiovisual translation (AVT) in the aftermath of the COVID-19 emergency, highlighting new trends with regards to the implementation of AI technologies as well as their strengths, constraints, and ethical implications. It starts with an overview of the current AVT landscape, focusing on future projections about its evolution and its critical aspects such as the worsening working conditions lamented by AVT professionals – especially freelancers – in recent years and how they might be affected by the advent of AI technologies in the industry. The second chapter delves into the history and development of three AI technologies which are used in combination with neural machine translation in automatic AVT tools: automatic speech recognition, speech synthesis and deepfakes (voice cloning and visual deepfakes for lip syncing), including real examples of start-up companies that utilize them – or are planning to do so – to localize audiovisual content automatically or semi-automatically. The third chapter explores the many ethical concerns around these innovative technologies, which extend far beyond the field of translation; at the same time, it attempts to revindicate their potential to bring about immense progress in terms of accessibility and international cooperation, provided that their use is properly regulated. Lastly, the fourth chapter describes two experiments, testing the efficacy of the currently available tools for automatic subtitling and automatic dubbing respectively, in order to take a closer look at their perks and limitations compared to more traditional approaches. This analysis aims to help discerning legitimate concerns from unfounded speculations with regards to the AI technologies which are entering the field of AVT; the intention behind it is to humbly suggest a constructive and optimistic view of the technological transformations that appear to be underway, whilst also acknowledging their potential risks.

Groupme: city data and multi-preference activity allocation

Classic group recommender systems focus on providing suggestions for a fixed group of people. Our work tries to give an inside look at design- ing a new recommender system that is capable of making suggestions for a sequence of activities, dividing people in subgroups, in order to boost over- all group satisfaction. However, this idea increases problem complexity in more dimensions and creates great challenge to the algorithm’s performance. To understand the e↵ectiveness, due to the enhanced complexity and pre- cise problem solving, we implemented an experimental system from data collected from a variety of web services concerning the city of Paris. The sys- tem recommends activities to a group of users from two di↵erent approaches: Local Search and Constraint Programming. The general results show that the number of subgroups can significantly influence the Constraint Program- ming Approaches’s computational time and e�cacy. Generally, Local Search can find results much quicker than Constraint Programming. Over a lengthy period of time, Local Search performs better than Constraint Programming, with similar final results.

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Un’applicazione di elaborazione del linguaggio naturale per il gioco di "Indovina chi?" su robot umanoide

La tesi è stata incentrata sul gioco «Indovina chi?» per l’identificazione da parte del robot Nao di un personaggio tramite la sua descrizione. In particolare la descrizione avviene tramite domande e risposte L’obiettivo della tesi è la progettazione di un sistema in grado di capire ed elaborare dei dati comunicati usando un sottoinsieme del linguaggio naturale, estrapolarne le informazioni chiave e ottenere un riscontro con informazioni date in precedenza. Si è quindi programmato il robot Nao in modo che sia in grado di giocare una partita di «Indovina chi?» contro un umano comunicando tramite il linguaggio naturale. Sono state implementate regole di estrazione e categorizzazione per la comprensione del testo utilizzando Cogito, una tecnologia brevettata dall'azienda Expert System. In questo modo il robot è in grado di capire le risposte e rispondere alle domande formulate dall'umano mediante il linguaggio naturale. Per il riconoscimento vocale è stata utilizzata l'API di Google e PyAudio per l'utilizzo del microfono. Il programma è stato implementato in Python e i dati dei personaggi sono memorizzati in un database che viene interrogato e modificato dal robot. L'algoritmo del gioco si basa su calcoli probabilistici di vittoria del robot e sulla scelta delle domande da proporre in base alle risposte precedentemente ricevute dall'umano. Le regole semantiche realizzate danno la possibilità al giocatore di formulare frasi utilizzando il linguaggio naturale, inoltre il robot è in grado di distinguere le informazioni che riguardano il personaggio da indovinare senza farsi ingannare. La percentuale di vittoria del robot ottenuta giocando 20 partite è stata del 50%. Il data base è stato sviluppato in modo da poter realizzare un identikit completo di una persona, oltre a quello dei personaggi del gioco. È quindi possibile ampliare il progetto per altri scopi, oltre a quello del gioco, nel campo dell'identificazione.

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.

Intermediate linguistic task fine-tuning on Multi-lingual models

State-of-the-art NLP systems are generally based on the assumption that the underlying models are provided with vast datasets to train on. However, especially when working in multi-lingual contexts, datasets are often scarce, thus more research should be carried out in this field. This thesis investigates the benefits of introducing an additional training step when fine-tuning NLP models, named Intermediate Training, which could be exploited to augment the data used for the training phase. The Intermediate Training step is applied by training models on NLP tasks that are not strictly related to the target task, aiming to verify if the models are able to leverage the learned knowledge of such tasks. Furthermore, in order to better analyze the synergies between different categories of NLP tasks, experimentations have been extended also to Multi-Task Training, in which the model is trained on multiple tasks at the same time.

Variational Quantum Splines: Moving Beyond Linearity for Quantum Activation Functions

Activation functions within neural networks play a crucial role in Deep Learning since they allow to learn complex and non-trivial patterns in the data. However, the ability to approximate non-linear functions is a significant limitation when implementing neural networks in a quantum computer to solve typical machine learning tasks. The main burden lies in the unitarity constraint of quantum operators, which forbids non-linearity and poses a considerable obstacle to developing such non-linear functions in a quantum setting. Nevertheless, several attempts have been made to tackle the realization of the quantum activation function in the literature. Recently, the idea of the QSplines has been proposed to approximate a non-linear activation function by implementing the quantum version of the spline functions. Yet, QSplines suffers from various drawbacks. Firstly, the final function estimation requires a post-processing step; thus, the value of the activation function is not available directly as a quantum state. Secondly, QSplines need many error-corrected qubits and a very long quantum circuits to be executed. These constraints do not allow the adoption of the QSplines on near-term quantum devices and limit their generalization capabilities. This thesis aims to overcome these limitations by leveraging hybrid quantum-classical computation. In particular, a few different methods for Variational Quantum Splines are proposed and implemented, to pave the way for the development of complete quantum activation functions and unlock the full potential of quantum neural networks in the field of quantum machine learning.

Motor Unit Spike Trains Recontruction from sEMG Signals for Gesture Classification on a Low-Power Processing Platform

Hand gesture recognition based on surface electromyography (sEMG) signals is a promising approach for the development of intuitive human-machine interfaces (HMIs) in domains such as robotics and prosthetics. The sEMG signal arises from the muscles' electrical activity, and can thus be used to recognize hand gestures. The decoding from sEMG signals to actual control signals is non-trivial; typically, control systems map sEMG patterns into a set of gestures using machine learning, failing to incorporate any physiological insight. This master thesis aims at developing a bio-inspired hand gesture recognition system based on neuromuscular spike extraction rather than on simple pattern recognition. The system relies on a decomposition algorithm based on independent component analysis (ICA) that decomposes the sEMG signal into its constituent motor unit spike trains, which are then forwarded to a machine learning classifier. Since ICA does not guarantee a consistent motor unit ordering across different sessions, 3 approaches are proposed: 2 ordering criteria based on firing rate and negative entropy, and a re-calibration approach that allows the decomposition model to retain information about previous sessions. Using a multilayer perceptron (MLP), the latter approach results in an accuracy up to 99.4% in a 1-subject, 1-degree of freedom scenario. Afterwards, the decomposition and classification pipeline for inference is parallelized and profiled on the PULP platform, achieving a latency < 50 ms and an energy consumption < 1 mJ. Both the classification models tested (a support vector machine and a lightweight MLP) yielded an accuracy > 92% in a 1-subject, 5-classes (4 gestures and rest) scenario. These results prove that the proposed system is suitable for real-time execution on embedded platforms and also capable of matching the accuracy of state-of-the-art approaches, while also giving some physiological insight on the neuromuscular spikes underlying the sEMG.

SmartForge: An event-driven architecture for simulator-based process optimization

In the metal industry, and more specifically in the forging one, scrap material is a crucial issue and reducing it would be an important goal to reach. Not only would this help the companies to be more environmentally friendly and more sustainable, but it also would reduce the use of energy and lower costs. At the same time, the techniques for Industry 4.0 and the advancements in Artificial Intelligence (AI), especially in the field of Deep Reinforcement Learning (DRL), may have an important role in helping to achieve this objective. This document presents the thesis work, a contribution to the SmartForge project, that was performed during a semester abroad at Karlstad University (Sweden). This project aims at solving the aforementioned problem with a business case of the company Bharat Forge Kilsta, located in Karlskoga (Sweden). The thesis work includes the design and later development of an event-driven architecture with microservices, to support the processing of data coming from sensors set up in the company's industrial plant, and eventually the implementation of an algorithm with DRL techniques to control the electrical power to use in it.

Mitigating non-Lambertian surfaces issues in Stereo Matching with Neural Radiance Fields

Depth estimation from images has long been regarded as a preferable alternative compared to expensive and intrusive active sensors, such as LiDAR and ToF. The topic has attracted the attention of an increasingly wide audience thanks to the great amount of application domains, such as autonomous driving, robotic navigation and 3D reconstruction. Among the various techniques employed for depth estimation, stereo matching is one of the most widespread, owing to its robustness, speed and simplicity in setup. Recent developments has been aided by the abundance of annotated stereo images, which granted to deep learning the opportunity to thrive in a research area where deep networks can reach state-of-the-art sub-pixel precision in most cases. Despite the recent findings, stereo matching still begets many open challenges, two among them being finding pixel correspondences in presence of objects that exhibits a non-Lambertian behaviour and processing high-resolution images. Recently, a novel dataset named Booster, which contains high-resolution stereo pairs featuring a large collection of labeled non-Lambertian objects, has been released. The work shown that training state-of-the-art deep neural network on such data improves the generalization capabilities of these networks also in presence of non-Lambertian surfaces. Regardless being a further step to tackle the aforementioned challenge, Booster includes a rather small number of annotated images, and thus cannot satisfy the intensive training requirements of deep learning. This thesis work aims to investigate novel view synthesis techniques to augment the Booster dataset, with ultimate goal of improving stereo matching reliability in presence of high-resolution images that displays non-Lambertian surfaces.

Query Library Development and Anomaly Detection on Flight Data

During the last semester of the Master’s Degree in Artificial Intelligence, I carried out my internship working for TXT e-Solution on the ADMITTED project. This paper describes the work done in those months. The thesis will be divided into two parts representing the two different tasks I was assigned during the course of my experience. The First part will be about the introduction of the project and the work done on the admittedly library, maintaining the code base and writing the test suits. The work carried out is more connected to the Software engineer role, developing features, fixing bugs and testing. The second part will describe the experiments done on the Anomaly detection task using a Deep Learning technique called Autoencoder, this task is on the other hand more connected to the data science role. The two tasks were not done simultaneously but were dealt with one after the other, which is why I preferred to divide them into two separate parts of this paper.

Hierarchical Multi-Label Text Classification in a Low-Resource Setting

In this thesis we address a multi-label hierarchical text classification problem in a low-resource setting and explore different approaches to identify the best one for our case. The goal is to train a model that classifies English school exercises according to a hierarchical taxonomy with few labeled data. The experiments made in this work employ different machine learning models and text representation techniques: CatBoost with tf-idf features, classifiers based on pre-trained models (mBERT, LASER), and SetFit, a framework for few-shot text classification. SetFit proved to be the most promising approach, achieving better performance when during training only a few labeled examples per class are available. However, this thesis does not consider all the hierarchical taxonomy, but only the first two levels: to address classification with the classes at the third level further experiments should be carried out, exploring methods for zero-shot text classification, data augmentation, and strategies to exploit the hierarchical structure of the taxonomy during training.

Voice conversion with pre-trained representations for audio anonymization

The recording and processing of voice data raises increasing privacy concerns for users and service providers. One way to address these issues is to move processing on the edge device closer to the recording so that potentially identifiable information is not transmitted over the internet. However, this is often not possible due to hardware limitations. An interesting alternative is the development of voice anonymization techniques that remove individual speakers characteristics while preserving linguistic and acoustic information in the data. In this work, a state-of-the-art approach to sequence-to-sequence speech conversion, ini- tially based on x-vectors and bottleneck features for automatic speech recognition, is explored to disentangle the two acoustic information using different pre-trained speech and speakers representation. Furthermore, different strategies for selecting target speech representations are analyzed. Results on public datasets in terms of equal error rate and word error rate show that good privacy is achieved with limited impact on converted speech quality relative to the original method.

Antibody-Antigen Interface Characterization with Contextualized Pre-trained Embeddings

Artificial Intelligence (AI) has substantially influenced numerous disciplines in recent years. Biology, chemistry, and bioinformatics are among them, with significant advances in protein structure prediction, paratope prediction, protein-protein interactions (PPIs), and antibody-antigen interactions. Understanding PPIs is critical since they are responsible for practically everything living and have several uses in vaccines, cancer, immunology, and inflammatory illnesses. Machine Learning (ML) offers enormous potential for effectively simulating antibody-antigen interactions and improving in-silico optimization of therapeutic antibodies for desired features, including binding activity, stability, and low immunogenicity. This research looks at the use of AI algorithms to better understand antibody-antigen interactions, and it further expands and explains several difficulties encountered in the field. Furthermore, we contribute by presenting a method that outperforms existing state-of-the-art strategies in paratope prediction from sequence data.