Offensive language detection is an ever-growing natural language processing (NLP) application. This growth is mainly because of the widespread usage of social networks, which becomes a mainstream channel for people to communicate, work, and enjoy entertainment content. Many incidents of sharing aggressive and offensive content negatively impacted society to a great extend. We believe contributing to improving and comparing different machine learning models to fight such harmful contents is an important and challenging goal for this thesis. We targeted the problem of offensive language detection for building efficient automated models for offensive language detection. With the recent advancements of NLP models, specifically, the Transformer model, which tackled many shortcomings of the standard seq-to-seq techniques. The BERT model has shown state-of-the-art results on many NLP tasks. Although the literature still exploring the reasons for the BERT achievements in the NLP field. Other efficient variants have been developed to improve upon the standard BERT, such as RoBERTa and ALBERT. Moreover, due to the multilingual nature of text on social media that could affect the model decision on a given tween, it is becoming essential to examine multilingual models such as XLM-RoBERTa trained on 100 languages and how did it compare to unilingual models. The RoBERTa based model proved to be the most capable model and achieved the highest F1 score for the tasks. Another critical aspect of a well-rounded offensive language detection system is the speed at which a model can be trained and make inferences. In that respect, we have considered the model run-time and fine-tuned the very efficient implementation of FastText called BlazingText that achieved good results, which is much faster than BERT-based models.

In this paper, we propose a novel learning-based polygonal point set tracking method. Compared to existing video object segmentation~(VOS) methods that propagate pixel-wise object mask information, we propagate a polygonal point set over frames. Specifically, the set is defined as a subset of points in the target contour, and our goal is to track corresponding points on the target contour. Those outputs enable us to apply various visual effects such as motion tracking, part deformation, and texture mapping. To this end, we propose a new method to track the corresponding points between frames by the global-local alignment with delicately designed losses and regularization terms. We also introduce a novel learning strategy using synthetic and VOS datasets that makes it possible to tackle the problem without developing the point correspondence dataset. Since the existing datasets are not suitable to validate our method, we build a new polygonal point set tracking dataset and demonstrate the superior performance of our method over the baselines and existing contour-based VOS methods. In addition, we present visual-effects applications of our method on part distortion and text mapping.

Automatically generating medical reports for retinal images is one of the promising ways to help ophthalmologists reduce their workload and improve work efficiency. In this work, we propose a new context-driven encoding network to automatically generate medical reports for retinal images. The proposed model is mainly composed of a multi-modal input encoder and a fused-feature decoder. Our experimental results show that our proposed method is capable of effectively leveraging the interactive information between the input image and context, i.e., keywords in our case. The proposed method creates more accurate and meaningful reports for retinal images than baseline models and achieves state-of-the-art performance. This performance is shown in several commonly used metrics for the medical report generation task: BLEU-avg (+16%), CIDEr (+10.2%), and ROUGE (+8.6%).

While pre-trained language models (e.g., BERT) have achieved impressive results on different natural language processing tasks, they have large numbers of parameters and suffer from big computational and memory costs, which make them difficult for real-world deployment. Therefore, model compression is necessary to reduce the computation and memory cost of pre-trained models. In this work, we aim to compress BERT and address the following two challenging practical issues: (1) The compression algorithm should be able to output multiple compressed models with different sizes and latencies, in order to support devices with different memory and latency limitations; (2) The algorithm should be downstream task agnostic, so that the compressed models are generally applicable for different downstream tasks. We leverage techniques in neural architecture search (NAS) and propose NAS-BERT, an efficient method for BERT compression. NAS-BERT trains a big supernet on a search space containing a variety of architectures and outputs multiple compressed models with adaptive sizes and latency. Furthermore, the training of NAS-BERT is conducted on standard self-supervised pre-training tasks (e.g., masked language model) and does not depend on specific downstream tasks. Thus, the compressed models can be used across various downstream tasks. The technical challenge of NAS-BERT is that training a big supernet on the pre-training task is extremely costly. We employ several techniques including block-wise search, search space pruning, and performance approximation to improve search efficiency and accuracy. Extensive experiments on GLUE and SQuAD benchmark datasets demonstrate that NAS-BERT can find lightweight models with better accuracy than previous approaches, and can be directly applied to different downstream tasks with adaptive model sizes for different requirements of memory or latency.

Bayesian neural networks and deep ensembles represent two modern paradigms of uncertainty quantification in deep learning. Yet these approaches struggle to scale mainly due to memory inefficiency issues, since they require parameter storage several times higher than their deterministic counterparts. To address this, we augment the weight matrix of each layer with a small number of inducing weights, thereby projecting the uncertainty quantification into such low dimensional spaces. We further extend Matheron's conditional Gaussian sampling rule to enable fast weight sampling, which enables our inference method to maintain reasonable run-time as compared with ensembles. Importantly, our approach achieves competitive performance to the state-of-the-art in prediction and uncertainty estimation tasks with fully connected neural networks and ResNets, while reducing the parameter size to $\leq 24.3\%$ of that of a $single$ neural network.

We introduce a novel inferential framework for marked point processes that enjoys both scalability and interpretability. The framework is based on variational inference and it aims to speed up inference for a flexible family of marked point processes where the joint distribution of times and marks can be specified in terms of the conditional distribution of times given the process filtration, and of the conditional distribution of marks given the process filtration and the current time. We assess the predictive ability of our proposed method over four real-world datasets where results show its competitive performance against other baselines. The attractiveness of our framework for the modelling of marked point processes is illustrated through a case study of association football data where scalability and interpretability are exploited for extracting useful informative patterns.

This paper introduces a new acquisition function under the Bayesian active learning framework, namely BABA. It is motivated by previously well-established works BALD, and BatchBALD which capture the mutual information between the model parameters and the predictive outputs of the data. Our proposed measure, BABA, endeavors to quantify the normalized mutual information by approximating the stochasticity of predictive probabilities using Beta distributions. BABA outperforms the well-known family of acquisition functions, including BALD and BatchBALD. We demonstrate this by showing extensive experimental results obtained from MNIST and EMNIST datasets.

The standard way to estimate the parameters $\Theta_\text{SEIR}$ (e.g., the transmission rate $\beta$) of an SEIR model is to use grid search, where simulations are performed on each set of parameters, and the parameter set leading to the least $L_2$ distance between predicted number of infections and observed infections is selected. This brute-force strategy is not only time consuming, as simulations are slow when the population is large, but also inaccurate, since it is impossible to enumerate all parameter combinations. To address these issues, in this paper, we propose to transform the non-differentiable problem of finding optimal $\Theta_\text{SEIR}$ to a differentiable one, where we first train a recurrent net to fit a small number of simulation data. Next, based on this recurrent net that is able to generalize SEIR simulations, we are able to transform the objective to a differentiable one with respect to $\Theta_\text{SEIR}$, and straightforwardly obtain its optimal value. The proposed strategy is both time efficient as it only relies on a small number of SEIR simulations, and accurate as we are able to find the optimal $\Theta_\text{SEIR}$ based on the differentiable objective. On two COVID-19 datasets, we observe that the proposed strategy leads to significantly better parameter estimations with a smaller number of simulations.

Machine learning and deep learning algorithms can be used to classify encrypted Internet traffic. Classification of encrypted traffic can become more challenging in the presence of adversarial attacks that target the learning algorithms. In this paper, we focus on investigating the effectiveness of different evasion attacks and see how resilient machine and deep learning algorithms are. Namely, we test C4.5 Decision Tree, K-Nearest Neighbor (KNN), Artificial Neural Network (ANN), Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). In most of our experimental results, deep learning shows better resilience against the adversarial samples in comparison to machine learning. Whereas, the impact of the attack varies depending on the type of attack.

In this article, we introduce a novel variant of the Tsetlin machine (TM) that randomly drops clauses, the key learning elements of a TM. In effect, TM with drop clause ignores a random selection of the clauses in each epoch, selected according to a predefined probability. In this way, additional stochasticity is introduced in the learning phase of TM. Along with producing more distinct and well-structured patterns that improve the performance, we also show that dropping clauses increases learning robustness. To explore the effects clause dropping has on accuracy, training time, and interpretability, we conduct extensive experiments on various benchmark datasets in natural language processing (NLP) (IMDb and SST2) as well as computer vision (MNIST and CIFAR10). In brief, we observe from +2% to +4% increase in accuracy and 2x to 4x faster learning. We further employ the Convolutional TM to document interpretable results on the CIFAR10 dataset. To the best of our knowledge, this is the first time an interpretable machine learning algorithm has been used to produce pixel-level human-interpretable results on CIFAR10. Also, unlike previous interpretable methods that focus on attention visualisation or gradient interpretability, we show that the TM is a more general interpretable method. That is, by producing rule-based propositional logic expressions that are \emph{human}-interpretable, the TM can explain how it classifies a particular instance at the pixel level for computer vision and at the word level for NLP.