The popularity of Artificial Intelligence (AI) technologies in various industries is increasing now more than ever before due to the ability of improving efficiency, enhancing decision-making and automating workflows. This demands that organizations need to be prepared to adopt these technologies to keep their competitive advantage and utilize the benefits in today's fast-paced business environment. There is a lack of guidance for organizations to adopt AI and further research of the organizational readiness factors is therefore needed to make sure the adoption of it is successful. The purpose of this research was to expand the knowledge of key factors that matter when organizations in the software industry want to create the best conditions before adopting the AI technologies in their business processes. The main contexts and factors were investigated with the technology-organizational-environmental (TOE) framework in synthesis with the technological readiness index (TRI) to get the perspective of both readiness and adoption. To answer the research questions that originated from the purpose, a qualitative research method was chosen where semi-structured interviews were conducted with managers with knowledge and experience in the field, as part of the empirical findings process. The most important contributing factor for readiness was communication, and the most obstructing factor was the discomfort of technological innovations such as AI. The main factors for a successful adoption were found to be the availability of slack resources and skilled labor and that the conditions of AI readiness are dealt with before adoption. The factor that could be classified as a main hindering factor in the adoption process was found to be a shortage of skilled labor in the market, with the right kind of knowledge and experience. / Populariteten för teknologier inom artificiell intelligens (AI) ökar nu mer än någonsin tidigare i olika branscher på grund av förmågan att förbättra effektiviteten, förbättra beslutsfattandet och möjligheten att automatisera arbetsflöden. Detta kräver att organisationer måste vara beredda att använda dessa teknologier för att behålla sina konkurrensfördelar och utnyttja fördelarna i dagens affärsmiljö där beslut fattas fort. Det finns dock en brist på vägledning för organisationer att ta till sig AI och ytterligare forskning om organisatoriska beredskaps faktorer behövs därför för att säkerställa att implementeringen av dessa teknologier blir framgångsrik. Syftet med denna forskning var att utöka kunskapen om nyckelfaktorer som verkligen betyder något när organisationer inom mjukvaruindustrin vill skapa de bästa förutsättningarna innan de tar till sig AI-teknologierna i sina affärsprocesser. De huvudsakliga sammanhangen och faktorerna undersöktes med hjälp av technology-organizational-environmental (TOE) ramverket i syntes med technological readiness index (TRI) för att få perspektiv på både beredskap och implementering av AI. För att besvara forskningsfrågorna valdes en kvalitativ forskningsmetod med semistrukturerade intervjuer för att samla in empirisk data. Dessa intervjuer genomfördes med chefer inom mjukvaruindustrin som hade erfarenhet kring implementering av AI. Den viktigaste bidragande faktorn för beredskapen var kommunikation, och den mest hindrande faktorn var obehaget för innovationer som AI. De huvudsakliga faktorerna för en framgångsrik implementering visade sig vara tillgången på överskotts resurser och kvalificerad arbetskraft och att villkoren för AI-beredskap hanteras innan implementering. Den faktor som kunde klassificeras som en huvudsaklig hämmande faktor i implementeringsprocessen visade sig vara brist på kvalificerad arbetskraft på marknaden, med rätt sorts kunskap och erfarenhet.

The rapid development of artificial intelligence (AI) has led to significant changes in the business environment and academic discussions. AI boosts productivity and positively impacts the competitive advantage of organisations. However, it also has its dark sides, such as prejudice, non-transparent processes, and people's fears that AI will be able to take their jobs in the future. The successful implementation of AI in organisations depends on several factors, including geopolitical, cultural, ecosystem, organisational, and individual factors. Geopolitical context and cultural differences can play an important role in the adoption and implementation of AI in organisations. This study examines the influence of geopolitical and cultural contexts on the decision-making process for the adoption and implementation of AI by managers from the fashion retail industry in Sweden and India. Given the extensive scope of these contexts, the authors narrowed their focus on specific factors. In the cultural context, the authors consider selected dimensions of the GLOBE project that reflect national culture. Within the Geopolitical context, particular attention is given to aspects such as data access and control, as well as the regulatory framework. In the course of this study, semi-structured interviews were conducted, and additional secondary data was studied. The study showed that the specifics of data access and control, as well as governmental legislative regulation, directly affect the decision-making process regarding the adoption and implementation of AI. As for the cultural context, here the degree of influence is heterogeneous, and decision-making on the implementation of AI is not always subject to the direct influence of the national cultural factors.

This master thesis focuses on the cutting-edge application of AI in developing intrusion detection systems (IDS) for unmanned aerial vehicles (UAVs) in smart cities. The objective is to address the escalating problem of UAV intrusions, which pose a significant risk to the safety and security of citizens and critical infrastructure. The thesis explores the current state of the art and provides a comprehensive understanding of recent advancements in the field, encompassing both physical and network attacks. The literature review examines various techniques and approaches employed in the development of AI-based IDS. This includes the utilization of machine learning algorithms, computer vision technologies, and edge computing. A proposed solution leveraging computer vision technologies is presented to detect and identify intruding UAVs in the sky effectively. The system employs machine learning algorithms to analyze video feeds from city-installed cameras, enabling real-time identification of potential intrusions. The proposed approach encompasses the detection of unauthorized drones, dangerous UAVs, and UAVs carrying suspicious payloads. Moreover, the thesis introduces a Cycle GAN network for image denoising that can translate noisy images to clean images without the need for paired training data. This approach employs two generators and two discriminators, incorporating a cycle consistency loss that ensures the generated images align with their corresponding input images. Furthermore, a distributed architecture is proposed for processing collected images using an edge-offloading approach within the UAV network. This architecture allows flying and ground cameras to leverage the computational capabilities of their IoT peers to process captured images. A hybrid neural network is developed to predict, based on input tasks, the potential edge computers capable of real-time processing. The edge-offloading approach reduces the computational burden on the centralized system and facilitates real-time analysis of network traffic, offering an efficient solution. In conclusion, the research outcomes of this thesis provide valuable insights into the development of secure and efficient IDS for UAVs in smart cities. The proposed solution contributes to the advancement of the UAV industry and enhances the safety and security of citizens and critical infrastructure within smart cities.

This thesis explores the potential of edge devices in three applications: indoor localization, urban traffic prediction, and multi-modal representation learning. For indoor localization, we propose a reliable data transmission network and robust data processing framework by visible light communications and machine learning to enhance the intelligence of smart buildings. The urban traffic prediction proposes a dynamic spatial and temporal origin-destination feature enhanced deep network with the graph convolutional network to collaboratively learn a low-dimensional representation for each region to predict in-traffic and out-traffic for every city region simultaneously. The multi-modal representation learning proposes using dynamic contexts to uniformly model visual and linguistic causalities, introducing a novel dynamic-contexts-based similarity metric that considers the correlation of potential causes and effects to measure the relevance among images. To enhance distributed training on edge devices, we introduced a new system called Distributed Artificial Intelligence Over-the-Air (AirDAI), which involves local training on raw data and sending trained outputs, such as model parameters, from local clients back to a central server for aggregation. To aid the development of AirDAI in wireless communication networks, we suggested a general system design and an associated simulator that can be tailored based on wireless channels and system-level configurations. We also conducted experiments to confirm the effectiveness and efficiency of the proposed system design and presented an analysis of the effects of wireless environments to facilitate future implementations and updates. This thesis proposes FedForest to address the communication and computation limitations in heterogeneous edge networks, which optimizes the global network by distilling knowledge from aggregated sub-networks. The sub-network sampling process is differentiable, and the model size is used as an additional constraint to extract a new sub-network for the subsequent local optimization process. FedForest significantly reduces server-to-client communication and local device computation costs compared to conventional algorithms while maintaining performance with the benchmark Top-K sparsification method. FedForest can accelerate the deployment of large-scale deep learning models on edge devices.

Modern machine learning techniques focus on extremely deep and multi-pathed networks, resulting in large memory and computational requirements. This thesis explores techniques for designing efficient convolutional networks including pixel shuffling, depthwise convolutions, and various activation fucntions. These techniques are then applied to two image processing domains: single-image super-resolution and image compression. The super-resolution model, TinyPSSR, is one-third the size of the next smallest model in literature while performing similar to or better than other larger models on representative test sets. The efficient deep image compression model is significantly smaller than any other model in literature and performs similarly in both computational cost and reconstruction quality to the JPEG standard.

While Artificial Intelligence (AI) is making giant steps, it is also raising concerns about its trustworthiness, due to the fact that widely-used black-box models cannot be exactly understood by humans. One of the ways to improve humans’ trust towards AI is to use interpretable AI models, i.e., models that can be thoroughly understood by humans, and thus trusted. However, interpretable AI models are not typically used in practice, as they are thought to be less performing than black-box models. This is more evident in Reinforce- ment Learning, where relatively little work addresses the problem of performing Reinforce- ment Learning with interpretable models. In this thesis, we address this gap, proposing methods for Interpretable Reinforcement Learning. For this purpose, we optimize Decision Trees by combining Reinforcement Learning with Evolutionary Computation techniques, which allows us to overcome some of the challenges tied to optimizing Decision Trees in Reinforcement Learning scenarios. The experimental results show that these approaches are competitive with the state-of-the-art score while being extremely easier to interpret. Finally, we show the practical importance of Interpretable AI by digging into the inner working of the solutions obtained.

AI-enabled information systems (AI-enabled IS) offer enhanced utility and efficiency due to their knowledge-based endowments, enabling human agents to assign and receive tasks from AI-enabled IS. As a result, this leads to improved decision-making, ability to manage laborious jobs, and a decrease in human errors. Despite the performance-based endowments and efficiencies, there are significant ethical concerns regarding the use of and delegation to AI-enabled IS, which have been extensively addressed in the literature on the dark side of artificial intelligence (AI). Notable concerns include bias and discrimination, fairness, transparency, privacy, accountability, and autonomy. However, the Information Systems (IS) literature does not have a delegation framework that incorporates ethics in the delegation mechanism. This work seeks to integrate a mixed deontological-teleological ethical system into the delegation mechanism to (and from) AI-enabled IS. To that end, I present a testable model to ethically appraise various AI-enabled IS as well as ethically evaluate delegation to (and from) AI-enabled IS in various settings and situations.

In recent years, generative AI has quickly become a new creative and artistic tool that could challenge our understanding of the creative process and the role of the machine. Despite having exhibited visually promising results, images generated by AI tools present various challenges, most notably their tendency to display cultural, gender and racial biases. The objective of the project is to speculate on the concept and prototype of an alternative text-to-image generation system, designed to mitigate biases from linguistic and cultural differences, and facilitate diversity in machine creativity. muGen, the final design, is a fictional system that allows the user to generate images using data in different languages, while adding user controls such as time period to better associate user’s idea with the system. / Under de senaste åren har generativ AI snabbt blivit ett nytt kreativt och konstnärligt verktyg som kan utmana vår förståelse av den kreativa processen och maskinens roll. Trots att bilder som genererats av AI-verktyg har uppvisat visuellt lovande resultat finns det flera utmaningar, framför allt deras tendens att visa kulturella, köns- och rasmässiga partiskhet. Syftet med projektet är att spekulera kring konceptet och prototypen för ett alternativt text-till-bild-genereringssystem, utformat för att mildra partiskhet från språkliga och kulturella skillnader, och underlätta mångfald i maskinkreativitet. muGen, den slutliga designen, är ett fiktivt system som låter användaren generera bilder med hjälp av data på olika språk, samtidigt som det lägger till användarkontroller som tidsperiod för att bättre associera användarens idé med systemet.

Thesis advisor: Jan Engelbrecht / Along with the proliferation of digital data collected using sensor technologies and a boost of computing power, Deep Learning (DL) based approaches have drawn enormous attention in the past decade due to their impressive performance in extracting complex relations from raw data and representing valuable information. At the same time, though, rooted in its notorious black-box nature, the appreciation of DL has been highly debated due to the lack of interpretability. On the one hand, DL only utilizes statistical features contained in raw data while ignoring human knowledge of the underlying system, which results in both data inefficiency and trust issues; on the other hand, a trained DL model does not provide to researchers any extra insight about the underlying system beyond its output, which, however, is the essence of most fields of science, e.g. physics and economics. The interpretability issue, in fact, has been naturally addressed in physics research. Conventional physics theories develop models of matter to describe experimentally observed phenomena. Tasks in DL, instead, can be considered as developing models of information to match with collected datasets. Motivated by techniques and perspectives in conventional physics, this thesis addresses the issue of interpretability in general information modeling. This thesis endeavors to address the two drawbacks of DL approaches mentioned above. Firstly, instead of relying on an intuition-driven construction of model structures, a problem-oriented perspective is applied to incorporate knowledge into modeling practice, where interesting mathematical properties emerge naturally which cast constraints on modeling. Secondly, given a trained model, various methods could be applied to extract further insights about the underlying system, which is achieved either based on a simplified function approximation of the complex neural network model, or through analyzing the model itself as an effective representation of the system. These two pathways are termed as guided model design (GuiMoD) and secondary measurements, respectively, which, together, present a comprehensive framework to investigate the general field of interpretability in modern Deep Learning practice. Remarkably, during the study of GuiMoD, a novel scheme emerges for the modeling practice in statistical learning: Algebraic Learning (AgLr). Instead of being restricted to the discussion of any specific model structure or dataset, AgLr starts from idiosyncrasies of a learning task itself and studies the structure of a legitimate model class in general. This novel modeling scheme demonstrates the noteworthy value of abstract algebra for general artificial intelligence, which has been overlooked in recent progress, and could shed further light on interpretable information modeling by offering practical insights from a formal yet useful perspective. / Thesis (PhD) — Boston College, 2021. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Physics.

Att ständigt behöva ändra sitt arbetssätt är något som är vanligt inom IT-branschen. Genom att anpassa sig till artificiell intelligens (AI) kan företag och individer dra nytta av fördelarna med tekniken. Däremot finns det en del utmaningar som är centrala att känna till. Syftet med studien är att undersöka hur programmeringsyrken potentiellt kommer att förändras och vilka utmaningar som behöver bemötas under utvecklingen. Genom en kvalitativ metod, intervjuades sex respondenter med varierande bakgrund och yrkesroller inom programmering. I syfte att undersöka deras perspektiv på AI-verktygens utveckling för att möjliggöra jämförelser gentemot teorin under olika teman som uppstod. Resultatet från studien visar på en osäkerhet kring hur AI-verktyg får användas, hur det genererade materialet får tillämpas och delade åsikter kring arbetsmarknadens utveckling. Vidare ser vi att företag och arbetstagare använder AI-verktyg med en viss försiktighet på grund av att juridiken kring AI inte är helt fastställd. Däremot visar resultatet att AI-verktygen effektiviserar arbetet för programmerare och kan användas som kreativ stöttning i utvecklingsprocessen. Rekommendationer för vidare studier inkluderar att undersöka ett specifikt yrke, användningen av ett specifikt AI-verktyg eller att se på skillnader utifrån använt programspråk. / Constantly needing to change one's way of working is common in the IT industry. By adapting to artificial intelligence (AI), companies and individuals can take advantage of the benefits of the technology. However, there are some challenges that are important to be aware of. The purpose of this study is to investigate how programming professions may potentially change and what challenges need to be addressed during this development. Using a qualitative method, six respondents with varying backgrounds and professional roles in programming were interviewed to examine their perspectives on the development of AI tools, to enable comparisons with theory under different themes that emerged. The results of the study show uncertainty about how AI tools can be used, how the generated material can be applied, and divided opinions on the development of the job market. Furthermore, we see that companies and employees use AI tools with some caution due to the fact that the legal aspects of AI are interpreted as a gray area. However, the results show that AI tools streamline the work of programmers and can be used as creative support in the development process. Recommendations for further studies include investigating a specific profession, the use of a specific AI tool, or examining differences based on the programming language used. The following essay is written in Swedish.