This paper intends to address the challenge of personalized recipe recommendation in the realm of diverse culinary preferences. The problem domain involves recipe recommendations, utilizing techniques such as association analysis and classification. Association analysis explores the relationships and connections between different ingredients to enhance the user experience. Meanwhile, the classification aspect involves categorizing recipes based on user-defined ingredients and preferences. A unique aspect of the paper is the consideration of recipes and ingredients belonging to multiple classes, recognizing the complexity of culinary combinations. This necessitates a sophisticated approach to classification and recommendation, ensuring the system accommodates the nature of recipe categorization. The paper seeks not only to recommend recipes but also to explore the process involved in achieving accurate and personalized recommendations.

Real-world tabular learning production scenarios typically involve evolving data streams, where data arrives continuously and its distribution may change over time. In such a setting, most studies in the literature regarding supervised learning favor the use of instance incremental algorithms due to their ability to adapt to changes in the data distribution. Another significant reason for choosing these algorithms is \textit{avoid storing observations in memory} as commonly done in batch incremental settings. However, the design of instance incremental algorithms often assumes immediate availability of labels, which is an optimistic assumption. In many real-world scenarios, such as fraud detection or credit scoring, labels may be delayed. Consequently, batch incremental algorithms are widely used in many real-world tasks. This raises an important question: "In delayed settings, is instance incremental learning the best option regarding predictive performance and computational efficiency?" Unfortunately, this question has not been studied in depth, probably due to the scarcity of real datasets containing delayed information. In this study, we conduct a comprehensive empirical evaluation and analysis of this question using a real-world fraud detection problem and commonly used generated datasets. Our findings indicate that instance incremental learning is not the superior option, considering on one side state-of-the-art models such as Adaptive Random Forest (ARF) and other side batch learning models such as XGBoost. Additionally, when considering the interpretability of the learning systems, batch incremental solutions tend to be favored. Code: \url{https://github.com/anselmeamekoe/DelayedLabelStream}

This thesis investigates the effectiveness of SimCLR, a contrastive learning technique, in Greek letter recognition, focusing on the impact of various augmentation techniques. We pretrain the SimCLR backbone using the Alpub dataset (pretraining dataset) and fine-tune it on a smaller ICDAR dataset (finetuning dataset) to compare SimCLR's performance against traditional baseline models, which use cross-entropy and triplet loss functions. Additionally, we explore the role of different data augmentation strategies, essential for the SimCLR training process. Methodologically, we examine three primary approaches: (1) a baseline model using cross-entropy loss, (2) a triplet embedding model with a classification layer, and (3) a SimCLR pretrained model with a classification layer. Initially, we train the baseline, triplet, and SimCLR models using 93 augmentations on ResNet-18 and ResNet-50 networks with the ICDAR dataset. From these, the top four augmentations are selected using a statistical t-test. Pretraining of SimCLR is conducted on the Alpub dataset, followed by fine-tuning on the ICDAR dataset. The triplet loss model undergoes a similar process, being pretrained on the top four augmentations before fine-tuning on ICDAR. Our experiments show that SimCLR does not outperform the baselines in letter recognition tasks. The baseline model with cross-entropy loss demonstrates better performance than both SimCLR and the triplet loss model. This study provides a detailed evaluation of contrastive learning for letter recognition, highlighting SimCLR's limitations while emphasizing the strengths of traditional supervised learning models in this task. We believe SimCLR's cropping strategies may cause a semantic shift in the input image, reducing training effectiveness despite the large pretraining dataset. Our code is available at https://github.com/DIVA-DIA/MT_augmentation_and_contrastive_learning/.

Recent advancements in unsupervised domain adaptation (UDA) and semi-supervised learning (SSL), particularly incorporating causality, have led to significant methodological improvements in these learning problems. However, a formal theory that explains the role of causality in the generalization performance of UDA/SSL is still lacking. In this paper, we consider the UDA/SSL scenarios where we access $m$ labelled source data and $n$ unlabelled target data as training instances under different causal settings with a parametric probabilistic model. We study the learning performance (e.g., excess risk) of prediction in the target domain from an information-theoretic perspective. Specifically, we distinguish two scenarios: the learning problem is called causal learning if the feature is the cause and the label is the effect, and is called anti-causal learning otherwise. We show that in causal learning, the excess risk depends on the size of the source sample at a rate of $O(\frac{1}{m})$ only if the labelling distribution between the source and target domains remains unchanged. In anti-causal learning, we show that the unlabelled data dominate the performance at a rate of typically $O(\frac{1}{n})$. These results bring out the relationship between the data sample size and the hardness of the learning problem with different causal mechanisms.

About one century ago, quantitative analysis and forecasting services were made available to businesses and the general public (Dominguez et al., 1988). Irving Fisher (1867-1947) spearheaded this initiative, pioneering data-driven forecasting practices grounded in theoretical foundations. His renowned theories, namely the monetary quantity theory, inflation-deflation theory, and index theory, were supported by statistical evidence. For example, Fisher (1925) was one of the earliest attempts to understand the source of business cycles in association with the price level (inflation), and Fisher (1926) was the first to report "a statistical relation between unemployment and price changes", today better known as the Phillips curve. Statistical inference is crucial for integrating economic theory and empirical data, which has become a tradition upheld by successive generations of researchers. In today's era of big data, we have unprecedented access to a vast amount of digital information about the economy. Recent advancements in statistical inference have uncovered new empirical patterns in prediction practices using datasets with temporal features. This paper aims at a plain quest: in a high dimensional linear predictive regression model when the number of potential regressors is larger than the sample size, how to conduct asymptotically valid statistical inference for a regressor of particular interest. To the best of our knowledge, no paper has solved this question before.

While achieving high prediction accuracy is a fundamental goal in machine learning, an equally important task is finding a small number of features with high explanatory power. One popular selection technique is permutation importance, which assesses a variable's impact by measuring the change in prediction error after permuting the variable. However, this can be problematic due to the need to create artificial data, a problem shared by other methods as well. Another problem is that variable selection methods can be limited by being model-specific. We introduce a new model-independent approach, Variable Priority (VarPro), which works by utilizing rules without the need to generate artificial data or evaluate prediction error. The method is relatively easy to use, requiring only the calculation of sample averages of simple statistics, and can be applied to many data settings, including regression, classification, and survival. We investigate the asymptotic properties of VarPro and show, among other things, that VarPro has a consistent filtering property for noise variables. Empirical studies using synthetic and real-world data show the method achieves a balanced performance and compares favorably to many state-of-the-art procedures currently used for variable selection.

In the realm of digital music, using tags to efficiently organize and retrieve music from extensive databases is crucial for music catalog owners. Human tagging by experts is labor-intensive but mostly accurate, whereas automatic tagging through supervised learning has approached satisfying accuracy but is restricted to a predefined set of training tags. Few-shot learning offers a viable solution to expand beyond this small set of predefined tags by enabling models to learn from only a few human-provided examples to understand tag meanings and subsequently apply these tags autonomously. We propose to integrate few-shot learning methodology into multi-label music auto-tagging by using features from pre-trained models as inputs to a lightweight linear classifier, also known as a linear probe. We investigate different popular pre-trained features, as well as different few-shot parametrizations with varying numbers of classes and samples per class. Our experiments demonstrate that a simple model with pre-trained features can achieve performance close to state-of-the-art models while using significantly less training data, such as 20 samples per tag. Additionally, our linear probe performs competitively with leading models when trained on the entire training dataset. The results show that this transfer learning-based few-shot approach could effectively address the issue of automatically assigning long-tail tags with only limited labeled data.

Multivariate time series data suffer from the problem of missing values, which hinders the application of many analytical methods. To achieve the accurate imputation of these missing values, exploiting inter-correlation by employing the relationships between sequences (i.e., a network) is as important as the use of temporal dependency, since a sequence normally correlates with other sequences. Moreover, exploiting an adequate network depending on time is also necessary since the network varies over time. However, in real-world scenarios, we normally know neither the network structure nor when the network changes beforehand. Here, we propose a missing value imputation method for multivariate time series, namely MissNet, that is designed to exploit temporal dependency with a state-space model and inter-correlation by switching sparse networks. The network encodes conditional independence between features, which helps us understand the important relationships for imputation visually. Our algorithm, which scales linearly with reference to the length of the data, alternatively infers networks and fills in missing values using the networks while discovering the switching of the networks. Extensive experiments demonstrate that MissNet outperforms the state-of-the-art algorithms for multivariate time series imputation and provides interpretable results.

Numerical simulation is powerful to study nonlinear solid mechanics problems. However, mesh-based or particle-based numerical methods suffer from the common shortcoming of being time-consuming, particularly for complex problems with real-time analysis requirements. This study presents a clustering adaptive Gaussian process regression (CAG) method aiming for real-time prediction for nonlinear structural responses in solid mechanics. It is a data-driven machine learning method featuring a small sample size, high accuracy, and high efficiency, leveraging nonlinear structural response patterns. Similar to the traditional Gaussian process regression (GPR) method, it operates in offline and online stages. In the offline stage, an adaptive sample generation technique is introduced to cluster datasets into distinct patterns for demand-driven sample allocation. This ensures comprehensive coverage of the critical samples for the solution space of interest. In the online stage, following the divide-and-conquer strategy, a pre-prediction classification categorizes problems into predefined patterns sequentially predicted by the trained multi-pattern Gaussian process regressor. In addition, dimension reduction and restoration techniques are employed in the proposed method to enhance its efficiency. A set of problems involving material, geometric, and boundary condition nonlinearities is presented to demonstrate the CAG method's abilities. The proposed method can offer predictions within a second and attain high precision with only about 20 samples within the context of this study, outperforming the traditional GPR using uniformly distributed samples for error reductions ranging from 1 to 3 orders of magnitude. The CAG method is expected to offer a powerful tool for real-time prediction of nonlinear solid mechanical problems and shed light on the complex nonlinear structural response pattern.

It is challenging to train a good intent classifier for a task-oriented dialogue system with only a few annotations. Recent studies have shown that fine-tuning pre-trained language models with a small amount of labeled utterances from public benchmarks in a supervised manner is extremely helpful. However, we find that supervised pre-training yields an anisotropic feature space, which may suppress the expressive power of the semantic representations. Inspired by recent research in isotropization, we propose to improve supervised pre-training by regularizing the feature space towards isotropy. We propose two regularizers based on contrastive learning and correlation matrix respectively, and demonstrate their effectiveness through extensive experiments. Our main finding is that it is promising to regularize supervised pre-training with isotropization to further improve the performance of few-shot intent detection. The source code can be found at https://github.com/fanolabs/isoIntentBert-main.