Relations that hold between causes and their effects are fundamental for a wide range of different sectors. Automatically finding sentences that express such relations may for example be of great interest for the economy or political institutions. However, for many languages other than English, a lack of training resources for this task needs to be dealt with. In recent years, large, pretrained transformer-based model architectures have proven to be very effective for tasks involving cross-lingual transfer such as cross-lingual language inference, as well as multilingual named entity recognition, POS-tagging and dependency parsing, which may hint at similar potentials for causality detection. In this thesis, we define causality detection as a binary labelling problem and use cross-lingual transfer to alleviate data scarcity for German and Swedish by using three different classifiers that make either use of multilingual sentence embeddings obtained from a pretrained encoder or pretrained multilingual language models. The source languages in most of our experiments will be English, for Swedish we however also use a small German training set and a combination of English and German training data.  We try out zero-shot transfer as well as making use of limited amounts of target language data either as a development set or as additional training data in a few-shot setting. In the latter scenario, we explore the impact of varying sizes of training data. Moreover, the problem of data scarcity in our situation also makes it necessary to work with data from different annotation projects. We also explore how much this would impact our result. For German as a target language, our results in a zero-shot scenario expectedly fall short in comparison with monolingual experiments, but F1-macro scores between 60 and 65 in cases where annotation did not differ drastically still signal that it was possible to transfer at least some knowledge. When introducing only small amounts of target language data, already notable improvements were observed and with the full German training data of about 3,000 sentences combined with the most suitable English data set, the performance for German in some scenarios even almost matches the state of the art for monolingual experiments on English. The best zero-shot performance on the Swedish data was even outperforming the scores achieved for German. However, due to problems with the additional Swedish training data, we were not able to improve upon the zero-shot performance in a few-shot setting in a similar manner as it was the case for German.

Multilingual natural language processing (NLP) is increasingly recognized for its potential in processing diverse text-type data, including those from social media, reviews, and technical reports. Multilingual language models like mBERT and XLM-RoBERTa (XLM-R) play a pivotal role in multilingual NLP. Notwithstanding their capabilities, the performance of these models largely relies on the availability of annotated training data. This thesis employs the multilingual pre-trained model XLM-R to examine its efficacy in sequence labelling to open-ended questions on democracy across multilingual surveys. Traditional annotation practices have been labour-intensive and time-consuming, with limited automation attempts. Previous studies often translated multilingual data into English, bypassing the challenges and nuances of native languages. Our study explores automatic multilingual annotation at the token level for democracy survey responses in five languages: Hungarian, Italian, Polish, Russian, and Spanish. The results reveal promising F1 scores, indicating the feasibility of using multilingual models for such tasks. However, the performance of these models is closely tied to the quality and nature of the training set. This research paves the way for future experiments and model adjustments, underscoring the importance of refining training data and optimizing model techniques for enhanced classification accuracy.

SOTA language models have demonstrated remarkable capabilities in tackling NLP tasks they have not been explicitly trained on – given a few demonstrations of the task (few-shot learning), or even none at all (zero-shot learning). The purpose of this Master’s thesis has been to investigate practical aspects and potential applications of zero/few-shot learning in the context of text classification. This includes topics such as combined usage with active learning, automated data labeling, and interpretability. Two different methods for zero/few-shot learning have been investigated, and the results indicate that:  • Active learning can be used to marginally improve few-shot performance, but it seems to be mostly beneficial in settings with very few samples (e.g. less than 10). • Zero-shot learning can be used produce reasonable candidate labels for classes in a dataset, given knowledge of the classification task at hand.  • It is difficult to trust the predictions of zero-shot text classification without access to a validation dataset, but IML methods such as saliency maps could find usage in debugging zero-shot models. / Ledande språkmodeller har uppvisat anmärkningsvärda förmågor i att lösa NLP-problem de inte blivit explicit tränade på – givet några exempel av problemet (few-shot learning), eller till och med inga alls (zero-shot learning). Syftet med det här examensarbetet har varit att undersöka praktiska aspekter och potentiella tillämpningar av zero/few-shot learning inom kontext av textklassificering. Detta inkluderar kombinerad användning med aktiv inlärning, automatiserad datamärkning, och tolkningsbarhet. Två olika metoder för zero/few-shot learning har undersökts, och resultaten indikerar att: • Aktiv inlärning kan användas för att marginellt förbättra textklassificering med few-shot learning, men detta verkar vara mest fördelaktigt i situationer med väldigt få datapunkter (t.ex. mindre än 10). • Zero-shot learning kan användas för att hitta lämpliga etiketter för klasser i ett dataset, givet kunskap om klassifikationsuppgiften av intresse. • Det är svårt att lita på robustheten i textklassificering med zero-shot learning utan tillgång till valideringsdata, men metoder inom tolkningsbar maskininlärning såsom saliency maps skulle kunna användas för att felsöka zero-shot modeller.

This Master's Thesis delves into the application of topological regularization techniques and relative latent representations within the realm of zero-shot model stitching. Building upon the prior work of Moschella et al. (2022) that introduces relative latent representations to enhance the similarities between latent spaces of different models, we incorporate the approach of Hofer et al. (2021), which combines Topological Data Analysis (TDA) and Machine Learning techniques for topological densification of class distributions in the latent space. The main research objective is to investigate the impact of topological regularization on zero-shot stitching performance when employing relative latent representations. Theoretical foundations for the relative transformation are established based on the intertwiner groups of activation functions. Empirical analyses are conducted to validate the assumptions underlying the construction of the relative transformation in the latent space. Moreover, experiments are performed on a Large Language Model trained on multilingual Amazon Reviews datasets to evaluate the effectiveness of zero-shot stitching while using the topological densification technique and the relative transformation. The findings indicate that the proposed methodologies can enhance the performance of multilingual model stitching. Specifically, enforcing the relative transformation to preserve the H0 homology death times distributions proves beneficial. Additionally, the presence of similar topological features plays a crucial role in achieving higher model compatibility. However, a more in-depth exploration of the geometric properties of the post-relative transformation latent space is necessary to further improve the topological densification technique. Overall, this work contributes to the emerging field of Topological Machine Learning and provides valuable insights for researchers in transfer learning and representation learning domains. / Denna masteruppsats undersöker tillämpningen av topologiska regleringstekniker och relativa latenta representationer inom området för zero-shot model stitching. Genom att bygga vidare på tidigare arbete av Moschella et al. (2022), som introducerade relativa latenta representationer för att förbättra likheterna mellan latenta rummet hos olika modeller, inkorporerar vi tillvägagångssättet av Hofer et al. (2021), som kombinerar topologisk dataanalys (TDA) och maskininlärningstekniker för topologisk ``förtätning'' av klassfördelningar i det latenta utrymmet. Den huvudsakliga forskningsuppgiften är att undersöka effekten av topologisk reglering på zero-shot model stitching-prestanda när man använder relativa latenta representationer. Teoretiska grunder för den relativa transformationen etableras baserat på intertwinergrupperna för aktiveringsfunktioner. Empiriska analyser genomförs för att validera antagandena som ligger till grund för konstruktionen av den relativa transformationen i det latenta rummen. Dessutom utförs experiment på en stor språkmodell tränad på multilinguella Amazon Reviews-dataset för att utvärdera effektiviteten hos zero-shot model stitching med Hofer's topologiska reglering och relativa transformation. Resultaten visar att de föreslagna metoderna kan förbättra prestationen hos zero-shot model stitching för flerspråkiga modeller. Specifikt är det fördelaktigt att tvinga den relativa transformationen att bevara H0 homologins dödstidsfördelningar. Dessutom spelar närvaron av liknande topologiska egenskaper en avgörande roll för att uppnå högre modellkompatibilitet. Dock krävs en mer ingående utforskning av de geometriska egenskaperna hos det latenta utrymmet efter den relativa transformationen för att ytterligare förbättra Hofer's topologiska reglering. Sammanfattningsvis bidrar detta arbete till det framväxande området Topologisk Maskininlärning och ger värdefulla insikter för forskare inom ``transfer-inlärning'' och representationsinlärningsdomäner.

There are many sustainability issues related to products, services, and business processes. For example, the production, usage, and disposal of IT equipment all impact sustainability. Therefore, buying more sustainable IT equipment can make a difference. More sustainable IT equipment can be acquired by selecting IT equipment with sustainability certification, such as TCO Certified. TCO Certified makes sustainable purchasing easier, which is useful in public procurement. Public procurement is complex to guarantee objectivity and transparency. Transparency in public procurement also means many public procurement documents are available, which can be analyzed. We hypothesized that the sustainability focuses in these public procurement documents (what the text is about) reflect the sustainability priorities of professional buyers (in their minds, which is indirectly observable). With this link, we investigated differences in sustainability priorities by using a machine learning model for predicting sustainability focuses in public procurement documents. By using a large language model, we automatically extracted sustainability focuses in procurement documents from the e-procurement platform TED. Thereby, we measured the sustainability focus of countries all over the globe. Through interviews with experts, we saw several indications that the used method is a good way of pinpointing sustainability priorities. We provide maps of sustainability focuses around the world (in section 4.12). Moreover, we analyze the results in-depth. One interesting finding includes indications that countries generally do not prioritize an issue more if the issue is of a larger concern. Counterintuitively, countries prioritize an issue more if the issue is of lesser concern. One example of this is circularity focus, which we note is generally less in countries with worse waste management. To our knowledge, analyzing sustainability focuses in procurement documents has not been done on this scale before. We believe these novel results can lead the way for a better understanding of sustainability priorities around the world. / Det finns många hållbarhetsproblem relaterade till produkter, tjänster och affärsprocesser. Till exempel finns det en hållbarhetspåverkan i produktion, användning och avfallshantering av IT-utrustning. Inköp av hållbarare IT-utrustning kan därför göra skillnad. Mer hållbar IT-utrustning kan erhållas genom att välja hållbarhetscertifierad IT-utrustning som exempelvis TCO Certified. TCO Certified gör hållbara inköp enklare och är särskilt användbart inom offentlig upphandling. Offentlig upphandling är komplext i objektivitet- och transparenssyfte. Transparens inom offentlig upphandling innebär också att många offentliga upphandlingsdokument finns tillgängliga och kan analyseras. Vi antog att hållbarhetsfokusen inuti dessa offentliga upphandlings-dokument (vad texten handlar om) hänger ihop med de professionella inköparnas hållbarhetsprioriteringar (deras tankar/känslor/intressen). Med denna länk undersökte vi skillnader i hållbarhetsprioriteringar genom att använda en maskininlärningsmodell för att uppskatta hållbarhetsfokus i offentliga upphandlingsdokument. Med en storskalig språkmodell extraherade vi automatiskt hållbarhetsfokus i dokument från Tenders Electronics Daily. Därigenom mätte vi hållbarhetsfokus för länder över hela världen. Genom intervjuer med experter såg vi flera indikationer på att den använda metoden är ett bra sätt att uppskatta hållbarhetsprioriteringar på. Vi presenterar kartor över hållbarhetsfokus runt om i världen (i avsnitt 4.12). Dessutom analyserar vi resultaten på djupet. En intressant upptäckt är att länder generellt inte prioriterar ett problem mer om problemet är allvarligare i det landet utan snarare tvärtom. Länder prioriterar ett problem mer om problemet är av lägre allvarlighetsgrad i det landet. Till exempel ser vi att cirkularitetsfokuset i offentliga upphandlingar är mindre i länder med sämre avfallshantering. Att analysera hållbarhetsfokus i upphandlingsdokument har inte tidigare gjorts på denna skala såvitt vi känner till. Vi tror våra resultat kan bidra till en bättre förståelse av hållbarhetsprioriteringar runt om i världen.

La construction d'algorithmes classifiant des images à grande échelle est devenue une tache essentielle du fait de la difficulté d'effectuer des recherches dans les immenses collections de données visuelles inetiquetées présentes sur Internet. Nous visons à classifier des images en fonction de leur contenu pour simplifier la gestion de telles bases de données. La classification d'images à grande échelle est un problème complèxe, de par l'importance de la taille des ensembles de données, tant en nombre d'images qu'en nombre de classes. Certaines de ces classes sont dites "fine-grained" (sémantiquement proches les unes des autres) et peuvent même ne contenir aucun représentant étiqueté. Dans cette thèse, nous utilisons des représentations état de l'art d'images et nous concentrons sur des méthodes d'apprentissage efficaces. Nos contributions sont (1) un banc d'essai d'algorithmes d'apprentissage pour la classification à grande échelle et (2) un nouvel algorithme basé sur l'incorporation d'étiquettes pour apprendre sur des données peu abondantes. En premier lieu, nous introduisons un banc d'essai d'algorithmes d'apprentissage pour la classification à grande échelle, dans le cadre entièrement supervisé. Il compare plusieurs fonctions objectifs pour apprendre des classifieurs linéaires, tels que "un contre tous", "multiclasse", "ranking", "ranking pondéré moyen" par descente de gradient stochastique. Ce banc d'essai se conclut en un ensemble de recommandations pour la classification à grande échelle. Avec une simple repondération des données, la stratégie "un contre tous" donne des performances meilleures que toutes les autres. Par ailleurs, en apprentissage en ligne, un pas d'apprentissage assez petit s'avère suffisant pour obtenir des résultats au niveau de l'état de l'art. Enfin, l'arrêt anticipé de la descente de gradient stochastique introduit une régularisation qui améliore la vitesse d'entraînement ainsi que la capacité de régularisation. Deuxièmement, face à des milliers de classes, il est parfois difficile de rassembler suffisamment de données d'entraînement pour chacune des classes. En particulier, certaines classes peuvent être entièrement dénuées d'exemples. En conséquence, nous proposons un nouvel algorithme adapté à ce scénario d'apprentissage dit "zero-shot". notre algorithme utilise des données parallèles, comme les attributs, pour incorporer les classes dans un espace euclidien. Nous introduisons par ailleurs une fonction pour mesurer la compatibilité entre image et étiquette. Les paramètres de cette fonction sont appris en utilisant un objectif de type "ranking". Notre algorithme dépasse l'état de l'art pour l'apprentissage "zero-shot", et fait preuve d'une grande flexibilité en permettant d'incorporer d'autres sources d'information parallèle, comme des hiérarchies. Il permet en outre une transition sans heurt du cas "zero-shot" au cas où peu d'exemples sont disponibles.

The mission of the research presented in this thesis is to give computers the power to sense and react to human activities. Without the ability to sense the surroundings and understand what humans are doing, computers will not be able to provide active, timely, appropriate, and considerate services to the humans. To accomplish this mission, the work stands on the shoulders of two giants: Machine learning and ubiquitous computing. Because of the ubiquity of sensor-enabled mobile and wearable devices, there has been an emerging opportunity to sense, learn, and infer human activities from the sensor data by leveraging state-of-the-art machine learning algorithms. While having shown promising results in human activity recognition, most existing approaches using supervised or semi-supervised learning have two fundamental problems. Firstly, most existing approaches require a large set of labeled sensor data for every target class, which requires a costly effort from human annotators. Secondly, an unseen new activity cannot be recognized if no training samples of that activity are available in the dataset. In light of these problems, a new approach in this area is proposed in our research. This thesis presents our novel approach to address the problem of human activity recognition when few or no training samples of the target activities are available. The main hypothesis is that the problem can be solved by the proposed NuActiv activity recognition framework, which consists of modeling the hierarchical and sequential structure of human activities, as well as bringing humans in the loop of model training. By injecting human knowledge about the hierarchical nature of human activities, a semantic attribute representation and a two-layer attribute-based learning approach are designed. To model the sequential structure, a probabilistic graphical model is further proposed to take into account the temporal dependency of activities and attributes. Finally, an active learning algorithm is developed to reinforce the recognition accuracy using minimal user feedback. The hypothesis and approaches presented in this thesis are validated by two case studies and real-world experiments on exercise activities and daily life activities. Experimental results show that the NuActiv framework can effectively recognize unseen new activities even without any training data, with up to 70-80% precision and recall rate. It also outperforms supervised learning with limited labeled data for the new classes. The results significantly advance the state of the art in human activity recognition, and represent a promising step towards bridging the gap between computers and humans.

Nous assistons actuellement à une explosion de la quantité des données visuelles. Par exemple, plusieurs millions de photos sont partagées quotidiennement sur les réseaux sociaux. Les méthodes d'interprétation d'images vise à faciliter l'accès à ces données visuelles, d'une manière sémantiquement compréhensible. Dans ce manuscrit, nous définissons certains buts détaillés qui sont intéressants pour les taches d'interprétation d'images, telles que la classification ou la recherche d'images, que nous considérons dans les trois chapitres principaux. Tout d'abord, nous visons l'exploitation de la nature multimodale de nombreuses bases de données, pour lesquelles les documents sont composés d'images et de descriptions textuelles. Dans ce but, nous définissons des similarités entre le contenu visuel d'un document, et la description textuelle d'un autre document. Ces similarités sont calculées en deux étapes, tout d'abord nous trouvons les voisins visuellement similaires dans la base multimodale, puis nous utilisons les descriptions textuelles de ces voisins afin de définir une similarité avec la description textuelle de n'importe quel document. Ensuite, nous présentons une série de modèles structurés pour la classification d'images, qui encodent explicitement les interactions binaires entre les étiquettes (ou labels). Ces modèles sont plus expressifs que des prédicateurs d'étiquette indépendants, et aboutissent à des prédictions plus fiables, en particulier dans un scenario de prédiction interactive, où les utilisateurs fournissent les valeurs de certaines des étiquettes d'images. Un scenario interactif comme celui-ci offre un compromis intéressant entre la précision, et l'effort d'annotation manuelle requis. Nous explorons les modèles structurés pour la classification multi-étiquette d'images, pour la classification d'image basée sur les attributs, et pour l'optimisation de certaines mesures de rang spécifiques. Enfin, nous explorons les classifieurs par k plus proches voisins, et les classifieurs par plus proche moyenne, pour la classification d'images à grande échelle. Nous proposons des méthodes d'apprentissage de métrique efficaces pour améliorer les performances de classification, et appliquons ces méthodes à une base de plus d'un million d'images d'apprentissage, et d'un millier de classes. Comme les deux méthodes de classification permettent d'incorporer des classes non vues pendant l'apprentissage à un coût presque nul, nous avons également étudié leur performance pour la généralisation. Nous montrons que la classification par plus proche moyenne généralise à partir d'un millier de classes, sur dix mille classes à un coût négligeable, et les performances obtenus sont comparables à l'état de l'art. / We are currently experiencing an exceptional growth of visual data, for example, millions of photos are shared daily on social-networks. Image understanding methods aim to facilitate access to this visual data in a semantically meaningful manner. In this dissertation, we define several detailed goals which are of interest for the image understanding tasks of image classification and retrieval, which we address in three main chapters. First, we aim to exploit the multi-modal nature of many databases, wherein documents consists of images with a form of textual description. In order to do so we define similarities between the visual content of one document and the textual description of another document. These similarities are computed in two steps, first we find the visually similar neighbors in the multi-modal database, and then use the textual descriptions of these neighbors to define a similarity to the textual description of any document. Second, we introduce a series of structured image classification models, which explicitly encode pairwise label interactions. These models are more expressive than independent label predictors, and lead to more accurate predictions. Especially in an interactive prediction scenario where a user provides the value of some of the image labels. Such an interactive scenario offers an interesting trade-off between accuracy and manual labeling effort. We explore structured models for multi-label image classification, for attribute-based image classification, and for optimizing for specific ranking measures. Finally, we explore k-nearest neighbors and nearest-class mean classifiers for large-scale image classification. We propose efficient metric learning methods to improve classification performance, and use these methods to learn on a data set of more than one million training images from one thousand classes. Since both classification methods allow for the incorporation of classes not seen during training at near-zero cost, we study their generalization performances. We show that the nearest-class mean classification method can generalize from one thousand to ten thousand classes at negligible cost, and still perform competitively with the state-of-the-art.

La construction d'algorithmes classifiant des images à grande échelle est devenue une t^ache essentielle du fait de la difficulté d'effectuer des recherches dans les immenses collections de données visuelles non-etiquetées présentes sur Internet. L'objetif est de classifier des images en fonction de leur contenu pour simplifier la gestion de telles bases de données. La classification d'images à grande échelle est un problème complexe, de par l'importance de la taille des ensembles de données, tant en nombre d'images qu'en nombre de classes. Certaines de ces classes sont dites "fine-grained" (sémantiquement proches les unes des autres) et peuvent même ne contenir aucun représentant étiqueté. Dans cette thèse, nous utilisons des représentations à l'état de l'art d'images et nous concentrons sur des méthodes d'apprentissage efficaces. Nos contributions sont (1) un banc d'essai d'algorithmes d'apprentissage pour la classification à grande échelle et (2) un nouvel algorithme basé sur l'incorporation d'étiquettes pour apprendre sur des données peu abondantes. En premier lieu, nous introduisons un banc d'essai d'algorithmes d'apprentissage pour la classification à grande échelle, dans un cadre entièrement supervisé. Il compare plusieurs fonctions objectifs pour apprendre des classifieurs linéaires, tels que "un contre tous", "multiclasse", "classement", "classement avec pondération" par descente de gradient stochastique. Ce banc d'essai se conclut en un ensemble de recommandations pour la classification à grande échelle. Avec une simple repondération des données, la stratégie "un contre tous" donne des performances meilleures que toutes les autres. Par ailleurs, en apprentissage en ligne, un pas d'apprentissage assez petit s'avère suffisant pour obtenir des résultats au niveau de l'état de l'art. Enfin, l'arrêt prématuré de la descente de gradient stochastique introduit une régularisation qui améliore la vitesse d'entraînement ainsi que la capacité de régularisation. Deuxièmement, face à des milliers de classes, il est parfois difficile de rassembler suffisamment de données d'entraînement pour chacune des classes. En particulier, certaines classes peuvent être entièrement dénuées d'exemples. En conséquence, nous proposons un nouvel algorithme adapté à ce scénario d'apprentissage dit "zero-shot". Notre algorithme utilise des données parallèles, comme les attributs, pour incorporer les classes dans un espace euclidien. Nous introduisons par ailleurs une fonction pour mesurer la compatibilité entre image et étiquette. Les paramètres de cette fonction sont appris en utilisant un objectif de type "ranking". Notre algorithme dépasse l'état de l'art pour l'apprentissage "zero-shot", et fait preuve d'une grande flexibilité en permettant d'incorporer d'autres sources d'information parallèle, comme des hiérarchies. Il permet en outre une transition sans heurt du cas "zero-shot" au cas où peu d'exemples sont disponibles. / Building algorithms that classify images on a large scale is an essential task due to the difficulty in searching massive amount of unlabeled visual data available on the Internet. We aim at classifying images based on their content to simplify the manageability of such large-scale collections. Large-scale image classification is a difficult problem as datasets are large with respect to both the number of images and the number of classes. Some of these classes are fine grained and they may not contain any labeled representatives. In this thesis, we use state-of-the-art image representations and focus on efficient learning methods. Our contributions are (1) a benchmark of learning algorithms for large scale image classification, and (2) a novel learning algorithm based on label embedding for learning with scarce training data. Firstly, we propose a benchmark of learning algorithms for large scale image classification in the fully supervised setting. It compares several objective functions for learning linear classifiers such as one-vs-rest, multiclass, ranking and weighted average ranking using the stochastic gradient descent optimization. The output of this benchmark is a set of recommendations for large-scale learning. We experimentally show that, online learning is well suited for large-scale image classification. With simple data rebalancing, One-vs-Rest performs better than all other methods. Moreover, in online learning, using a small enough step size with respect to the learning rate is sufficient for state-of-the-art performance. Finally, regularization through early stopping results in fast training and a good generalization performance. Secondly, when dealing with thousands of classes, it is difficult to collect sufficient labeled training data for each class. For some classes we might not even have a single training example. We propose a novel algorithm for this zero-shot learning scenario. Our algorithm uses side information, such as attributes to embed classes in a Euclidean space. We also introduce a function to measure the compatibility between an image and a label. The parameters of this function are learned using a ranking objective. Our algorithm outperforms the state-of-the-art for zero-shot learning. It is flexible and can accommodate other sources of side information such as hierarchies. It also allows for a smooth transition from zero-shot to few-shots learning.

[Background]: The advent of bidirectional encoder representation from trans- formers (BERT) language models (Devlin et al., 2018) and MS Marco, a large scale human-annotated dataset for machine reading comprehension (Bajaj et al., 2016) that made publicly available, led the field of information retrieval (IR) to experience a revolution (Lin et al., 2020). The retrieval model based on BERT of Nogueira and Cho (2019), by the time they published their paper, became the top entry in the MS Marco passage-reranking leaderboard, surpassing the previous state of the art by 27% in MRR@10. However, training such neural IR models for different domains than MS Marco is still hard because neural approaches often require a vast amount of training data to perform effectively, which is not always available. To address the problem of the shortage of labelled data a new line of research emerged, training neural models with weak supervision. In weak supervision, given an unlabelled dataset labels are generated automatically using an existing model and then a machine learning model is trained upon the artificial “weak“ data. In case of weak supervision for IR, the training dataset comes in the form of a tuple (query, passage). Dehghani et al. (2017) in their work used the AOL query logs (Pass et al., 2006), which is a set of millions of real web queries, and BM25 to retrieve the relevant passages for each of the user queries. A drawback with this approach is that it is hard to obtain query logs for every single different domain. [Objective]: This thesis proposes an intuitive approach for addressing the shortage of data in domains with limited or no data at all through transfer learning in the context of IR. We leverage Wikipedia’s structure for creating a Wikipedia-based generic IR training dataset for zero-shot neural models. [Method]: We create the “pseudo-queries“ by concatenating the titles of Wikipedia’s articles along with each of their title sections and we consider the associated section’s passage as the relevant passage of the pseudo-queries. All of our experiments are evaluated on a standard collection: MS Marco, which is a large scale web collection. For our zero-shot experiments, our proposed model, called “Wiki“, is a BERT model trained on the artificial Wikipedia-based dataset and the baseline is a default BERT model without any additional training. In our second line of experiments, we explore the benefits gained by pre-fine- tuning on the Wikipedia-based IR dataset and further fine-tuning on in-domain data. Our proposed model, "Wiki+Ma", is a BERT model pre-fine-tuned in the Wikipedia-based dataset and further fine-tuned in MS Marco, while the baseline is a BERT model fine-tuned only in MS Marco. [Results]: Results regarding our first experiments show that our BERT model trained on the Wikipedia-based IR dataset, called "Wiki", achieves a performance of 0.197 in MRR@10, which is about +10 points more in comparison to a BERT model with default weights; in addition, results in the development set indicate that the “Wiki“ model performs better than BERT model trained on in-domain data when the data is between 10k-50k instances. Results regarding our second line of experiments show that pre-fine-tuning on the Wikipedia-based IR dataset benefits later fine-tuning steps on in-domain data in terms of stability. [Conclusion]: Our findings suggest that transfer learning for IR tasks by leveraging the generic knowledge incorporated in Wikipedia is possible, though more experimentation is needed to understand its limitations in comparison with the traditional approaches such as the BM25.