Continual learning for computer vision applications

One of the most visionary goals of Artificial Intelligence is to create a system able to mimic and eventually surpass the intelligence observed in biological systems including, ambitiously, the one observed in humans. The main distinctive strength of humans is their ability to build a deep understanding of the world by learning continuously and drawing from their experiences. This ability, which is found in various degrees in all intelligent biological beings, allows them to adapt and properly react to changes by incrementally expanding and refining their knowledge. Arguably, achieving this ability is one of the main goals of Artificial Intelligence and a cornerstone towards the creation of intelligent artificial agents. Modern Deep Learning approaches allowed researchers and industries to achieve great advancements towards the resolution of many long-standing problems in areas like Computer Vision and Natural Language Processing. However, while this current age of renewed interest in AI allowed for the creation of extremely useful applications, a concerningly limited effort is being directed towards the design of systems able to learn continuously. The biggest problem that hinders an AI system from learning incrementally is the catastrophic forgetting phenomenon. This phenomenon, which was discovered in the 90s, naturally occurs in Deep Learning architectures where classic learning paradigms are applied when learning incrementally from a stream of experiences. This dissertation revolves around the Continual Learning field, a sub-field of Machine Learning research that has recently made a comeback following the renewed interest in Deep Learning approaches. This work will focus on a comprehensive view of continual learning by considering algorithmic, benchmarking, and applicative aspects of this field. This dissertation will also touch on community aspects such as the design and creation of research tools aimed at supporting Continual Learning research, and the theoretical and practical aspects concerning public competitions in this field.

Design, implementation and evaluation of a virtual laboratory for Computer Engineering Education

Broad consensus has been reached within the Education and Cognitive Psychology research communities on the need to center the learning process on experimentation and concrete application of knowledge, rather than on a bare transfer of notions. Several advantages arise from this educational approach, ranging from the reinforce of students learning, to the increased opportunity for a student to gain greater insight into the studied topics, up to the possibility for learners to acquire practical skills and long-lasting proficiency. This is especially true in Engineering education, where integrating conceptual knowledge and practical skills assumes a strategic importance. In this scenario, learners are called to play a primary role. They are actively involved in the construction of their own knowledge, instead of passively receiving it. As a result, traditional, teacher-centered learning environments should be replaced by novel learner-centered solutions. Information and Communication Technologies enable the development of innovative solutions that provide suitable answers to the need for the availability of experimentation supports in educational context. Virtual Laboratories, Adaptive Web-Based Educational Systems and Computer-Supported Collaborative Learning environments can significantly foster different learner-centered instructional strategies, offering the opportunity to enhance personalization, individualization and cooperation. More specifically, they allow students to explore different kinds of materials, to access and compare several information sources, to face real or realistic problems and to work on authentic and multi-facet case studies. In addition, they encourage cooperation among peers and provide support through coached and scaffolded activities aimed at fostering reflection and meta-cognitive reasoning. This dissertation will guide readers within this research field, presenting both the theoretical and applicative results of a research aimed at designing an open, flexible, learner-centered virtual lab for supporting students in learning Information Security.

Methods for integrating machine learning and constrained optimization

In the framework of industrial problems, the application of Constrained Optimization is known to have overall very good modeling capability and performance and stands as one of the most powerful, explored, and exploited tool to address prescriptive tasks. The number of applications is huge, ranging from logistics to transportation, packing, production, telecommunication, scheduling, and much more. The main reason behind this success is to be found in the remarkable effort put in the last decades by the OR community to develop realistic models and devise exact or approximate methods to solve the largest variety of constrained or combinatorial optimization problems, together with the spread of computational power and easily accessible OR software and resources. On the other hand, the technological advancements lead to a data wealth never seen before and increasingly push towards methods able to extract useful knowledge from them; among the data-driven methods, Machine Learning techniques appear to be one of the most promising, thanks to its successes in domains like Image Recognition, Natural Language Processes and playing games, but also the amount of research involved. The purpose of the present research is to study how Machine Learning and Constrained Optimization can be used together to achieve systems able to leverage the strengths of both methods: this would open the way to exploiting decades of research on resolution techniques for COPs and constructing models able to adapt and learn from available data. In the first part of this work, we survey the existing techniques and classify them according to the type, method, or scope of the integration; subsequently, we introduce a novel and general algorithm devised to inject knowledge into learning models through constraints, Moving Target. In the last part of the thesis, two applications stemming from real-world projects and done in collaboration with Optit will be presented.

Exploring domain-informed and physics-guided learning in image-to-image translation

Image-to-image (i2i) translation networks can generate fake images beneficial for many applications in augmented reality, computer graphics, and robotics. However, they require large scale datasets and high contextual understanding to be trained correctly. In this thesis, we propose strategies for solving these problems, improving performances of i2i translation networks by using domain- or physics-related priors. The thesis is divided into two parts. In Part I, we exploit human abstraction capabilities to identify existing relationships in images, thus defining domains that can be leveraged to improve data usage efficiency. We use additional domain-related information to train networks on web-crawled data, hallucinate scenarios unseen during training, and perform few-shot learning. In Part II, we instead rely on physics priors. First, we combine realistic physics-based rendering with generative networks to boost outputs realism and controllability. Then, we exploit naive physical guidance to drive a manifold reorganization, which allowed generating continuous conditions such as timelapses.

Computational creativity: an interdisciplinary approach to sequential learning and creative generations

Creativity seems mysterious; when we experience a creative spark, it is difficult to explain how we got that idea, and we often recall notions like ``inspiration" and ``intuition" when we try to explain the phenomenon. The fact that we are clueless about how a creative idea manifests itself does not necessarily imply that a scientific explanation cannot exist. We are unaware of how we perform certain tasks, such as biking or language understanding, but we have more and more computational techniques that can replicate and hopefully explain such activities. We should understand that every creative act is a fruit of experience, society, and culture. Nothing comes from nothing. Novel ideas are never utterly new; they stem from representations that are already in mind. Creativity involves establishing new relations between pieces of information we had already: then, the greater the knowledge, the greater the possibility of finding uncommon connections, and the more the potential to be creative. In this vein, a beneficial approach to a better understanding of creativity must include computational or mechanistic accounts of such inner procedures and the formation of the knowledge that enables such connections. That is the aim of Computational Creativity: to develop computational systems for emulating and studying creativity. Hence, this dissertation focuses on these two related research areas: discussing computational mechanisms to generate creative artifacts and describing some implicit cognitive processes that can form the basis for creative thoughts.

Concepts and methods for efficient decentralized learning in federated settings

Deep Neural Networks (DNNs) have revolutionized a wide range of applications beyond traditional machine learning and artificial intelligence fields, e.g., computer vision, healthcare, natural language processing and others. At the same time, edge devices have become central in our society, generating an unprecedented amount of data which could be used to train data-hungry models such as DNNs. However, the potentially sensitive or confidential nature of gathered data poses privacy concerns when storing and processing them in centralized locations. To this purpose, decentralized learning decouples model training from the need of directly accessing raw data, by alternating on-device training and periodic communications. The ability of distilling knowledge from decentralized data, however, comes at the cost of facing more challenging learning settings, such as coping with heterogeneous hardware and network connectivity, statistical diversity of data, and ensuring verifiable privacy guarantees. This Thesis proposes an extensive overview of decentralized learning literature, including a novel taxonomy and a detailed description of the most relevant system-level contributions in the related literature for privacy, communication efficiency, data and system heterogeneity, and poisoning defense. Next, this Thesis presents the design of an original solution to tackle communication efficiency and system heterogeneity, and empirically evaluates it on federated settings. For communication efficiency, an original method, specifically designed for Convolutional Neural Networks, is also described and evaluated against the state-of-the-art. Furthermore, this Thesis provides an in-depth review of recently proposed methods to tackle the performance degradation introduced by data heterogeneity, followed by empirical evaluations on challenging data distributions, highlighting strengths and possible weaknesses of the considered solutions. Finally, this Thesis presents a novel perspective on the usage of Knowledge Distillation as a mean for optimizing decentralized learning systems in settings characterized by data heterogeneity or system heterogeneity. Our vision on relevant future research directions close the manuscript.

Thermal and air quality modeling of an electric vehicle cabin with low-cost sensors and reinforcement learning

The design process of any electric vehicle system has to be oriented towards the best energy efficiency, together with the constraint of maintaining comfort in the vehicle cabin. Main aim of this study is to research the best thermal management solution in terms of HVAC efficiency without compromising occupant’s comfort and internal air quality. An Arduino controlled Low Cost System of Sensors was developed and compared against reference instrumentation (average R-squared of 0.92) and then used to characterise the vehicle cabin in real parking and driving conditions trials. Data on the energy use of the HVAC was retrieved from the car On-Board Diagnostic port. Energy savings using recirculation can reach 30 %, but pollutants concentration in the cabin builds up in this operating mode. Moreover, the temperature profile appeared strongly nonuniform with air temperature differences up to 10° C. Optimisation methods often require a high number of runs to find the optimal configuration of the system. Fast models proved to be beneficial for these task, while CFD-1D model are usually slower despite the higher level of detail provided. In this work, the collected dataset was used to train a fast ML model of both cabin and HVAC using linear regression. Average scaled RMSE over all trials is 0.4 %, while computation time is 0.0077 ms for each second of simulated time on a laptop computer. Finally, a reinforcement learning environment was built in OpenAI and Stable-Baselines3 using the built-in Proximal Policy Optimisation algorithm to update the policy and seek for the best compromise between comfort, air quality and energy reward terms. The learning curves show an oscillating behaviour overall, with only 2 experiments behaving as expected even if too slow. This result leaves large room for improvement, ranging from the reward function engineering to the expansion of the ML model.