Several decades ago the Proximal Point Algorithm (PPA) started to gain a long-lasting attraction for both abstract operator theory and numerical optimization communities. Even in modern applications, researchers still use proximal minimization theory to design scalable algorithms that overcome nonsmoothness. Remarkable works as \cite{Fer:91,Ber:82constrained,Ber:89parallel,Tom:11} established tight relations between the convergence behaviour of PPA and the regularity of the objective function. In this manuscript we derive nonasymptotic iteration complexity of exact and inexact PPA to minimize convex functions under $\gamma-$Holderian growth: $\BigO{\log(1/\epsilon)}$ (for $\gamma \in [1,2]$) and $\BigO{1/\epsilon^{\gamma - 2}}$ (for $\gamma > 2$). In particular, we recover well-known results on PPA: finite convergence for sharp minima and linear convergence for quadratic growth, even under presence of deterministic noise. Moreover, when a simple Proximal Subgradient Method is recurrently called as an inner routine for computing each IPPA iterate, novel computational complexity bounds are obtained for Restarting Inexact PPA. Our numerical tests show improvements over existing restarting versions of the Subgradient Method.

Recent studies revealed that convolutional neural networks do not generalize well to small image transformations, e.g. rotations by a few degrees or translations of a few pixels. To improve the robustness to such transformations, we propose to introduce data augmentation at intermediate layers of the neural architecture, in addition to the common data augmentation applied on the input images. By introducing small perturbations to activation maps (features) at various levels, we develop the capacity of the neural network to cope with such transformations. We conduct experiments on three image classification benchmarks (Tiny ImageNet, Caltech-256 and Food-101), considering two different convolutional architectures (ResNet-18 and DenseNet-121). When compared with two state-of-the-art stabilization methods, the empirical results show that our approach consistently attains the best trade-off between accuracy and mean flip rate.

An important preliminary step of optical character recognition systems is the detection of text rows. To address this task in the context of historical data with missing labels, we propose a self-paced learning algorithm capable of improving the row detection performance. We conjecture that pages with more ground-truth bounding boxes are less likely to have missing annotations. Based on this hypothesis, we sort the training examples in descending order with respect to the number of ground-truth bounding boxes, and organize them into k batches. Using our self-paced learning method, we train a row detector over k iterations, progressively adding batches with less ground-truth annotations. At each iteration, we combine the ground-truth bounding boxes with pseudo-bounding boxes (bounding boxes predicted by the model itself) using non-maximum suppression, and we include the resulting annotations at the next training iteration. We demonstrate that our self-paced learning strategy brings significant performance gains on two data sets of historical documents, improving the average precision of YOLOv4 with more than 12% on one data set and 39% on the other.

Large transformer-based pre-trained language models have achieved impressive performance on a variety of knowledge-intensive tasks and can capture factual knowledge in their parameters. We argue that storing large amounts of knowledge in the model parameters is sub-optimal given the ever-growing amounts of knowledge and resource requirements. We posit that a more efficient alternative is to provide explicit access to contextually relevant structured knowledge to the model and train it to use that knowledge. We present LM-CORE -- a general framework to achieve this -- that allows \textit{decoupling} of the language model training from the external knowledge source and allows the latter to be updated without affecting the already trained model. Experimental results show that LM-CORE, having access to external knowledge, achieves significant and robust outperformance over state-of-the-art knowledge-enhanced language models on knowledge probing tasks; can effectively handle knowledge updates; and performs well on two downstream tasks. We also present a thorough error analysis highlighting the successes and failures of LM-CORE.

Automatic identification of emotions expressed in Twitter data has a wide range of applications. We create a well-balanced dataset by adding a neutral class to a benchmark dataset consisting of four emotions: fear, sadness, joy, and anger. On this extended dataset, we investigate the use of Support Vector Machine (SVM) and Bidirectional Encoder Representations from Transformers (BERT) for emotion recognition. We propose a novel ensemble model by combining the two BERT and SVM models. Experiments show that the proposed model achieves a state-of-the-art accuracy of 0.91 on emotion recognition in tweets.

Autoencoders are composed of coding and decoding units, hence they hold the inherent potential of high-performance data compression and signal compressed sensing. The main disadvantages of current autoencoders comprise the following several aspects: the research objective is not data reconstruction but feature representation; the performance evaluation of data recovery is neglected; it is hard to achieve lossless data reconstruction by pure autoencoders, even by pure deep learning. This paper aims for image reconstruction of autoencoders, employs cascade decoders-based autoencoders, perfects the performance of image reconstruction, approaches gradually lossless image recovery, and provides solid theory and application basis for autoencoders-based image compression and compressed sensing. The proposed serial decoders-based autoencoders include the architectures of multi-level decoders and the related optimization algorithms. The cascade decoders consist of general decoders, residual decoders, adversarial decoders and their combinations. It is evaluated by the experimental results that the proposed autoencoders outperform the classical autoencoders in the performance of image reconstruction.

Closed-loop architecture is widely utilized in automatic control systems and attain distinguished performance. However, classical compressive sensing systems employ open-loop architecture with separated sampling and reconstruction units. Therefore, a method of iterative compensation recovery for image compressive sensing (ICRICS) is proposed by introducing closed-loop framework into traditional compresses sensing systems. The proposed method depends on any existing approaches and upgrades their reconstruction performance by adding negative feedback structure. Theory analysis on negative feedback of compressive sensing systems is performed. An approximate mathematical proof of the effectiveness of the proposed method is also provided. Simulation experiments on more than 3 image datasets show that the proposed method is superior to 10 competition approaches in reconstruction performance. The maximum increment of average peak signal-to-noise ratio is 4.36 dB and the maximum increment of average structural similarity is 0.034 on one dataset. The proposed method based on negative feedback mechanism can efficiently correct the recovery error in the existing systems of image compressive sensing.

Automatic detection of fake news is a highly important task in the contemporary world. This study reports the 2nd shared task called UrduFake@FIRE2021 on identifying fake news detection in Urdu. The goal of the shared task is to motivate the community to come up with efficient methods for solving this vital problem, particularly for the Urdu language. The task is posed as a binary classification problem to label a given news article as a real or a fake news article. The organizers provide a dataset comprising news in five domains: (i) Health, (ii) Sports, (iii) Showbiz, (iv) Technology, and (v) Business, split into training and testing sets. The training set contains 1300 annotated news articles -- 750 real news, 550 fake news, while the testing set contains 300 news articles -- 200 real, 100 fake news. 34 teams from 7 different countries (China, Egypt, Israel, India, Mexico, Pakistan, and UAE) registered to participate in the UrduFake@FIRE2021 shared task. Out of those, 18 teams submitted their experimental results, and 11 of those submitted their technical reports, which is substantially higher compared to the UrduFake shared task in 2020 when only 6 teams submitted their technical reports. The technical reports submitted by the participants demonstrated different data representation techniques ranging from count-based BoW features to word vector embeddings as well as the use of numerous machine learning algorithms ranging from traditional SVM to various neural network architectures including Transformers such as BERT and RoBERTa. In this year's competition, the best performing system obtained an F1-macro score of 0.679, which is lower than the past year's best result of 0.907 F1-macro. Admittedly, while training sets from the past and the current years overlap to a large extent, the testing set provided this year is completely different.

ABSTRACT Coronal mass ejections (CMEs) are the most geoeffective space weather phenomena, being associated with large geomagnetic storms, having the potential to cause disturbances to telecommunication, satellite network disruptions, power grid damages and failures. Thus, considering these storms' potential effects on human activities, accurate forecasts of the geoeffectiveness of CMEs are paramount. This work focuses on experimenting with different machine learning methods trained on white-light coronagraph datasets of close to sun CMEs, to estimate whether such a newly erupting ejection has the potential to induce geomagnetic activity. We developed binary classification models using logistic regression, K-Nearest Neighbors, Support Vector Machines, feed forward artificial neural networks, as well as ensemble models. At this time, we limited our forecast to exclusively use solar onset parameters, to ensure extended warning times. We discuss the main challenges of this task, namely the extreme imbalance between the number of geoeffective and ineffective events in our dataset, along with their numerous similarities and the limited number of available variables. We show that even in such conditions, adequate hit rates can be achieved with these models. INTRODUCTION The purpose of this work is to develop a machine learning (ML) based model that can predict whether a coronal mass ejection (CME) will be geoeffective, using only numerical solar parameters as input. Coronal mass ejections are solar eruptive events whose magnetically charged particles can, directly or indirectly, under certain circumstances, reach Earth and cause geomagnetic storms (GSs), i.e., be geoeffective. These storms represent perturbations in the Earth's magnetic field, which have the potential to lead to electrical systems and grids failure and/or damage, power outages, navigation errors, radio signal perturbations, significant exposure to dangerous radiations for astronauts during space missions, etc. Given the potential negative impacts of such storms, predicting their occurrence is paramount for enabling safeguarding of human technology (Schwenn 2006; Pulkkinen 2007; Council 2013; Vourlidas et al. 2019; Temmer 2021). The intensity of the storms can be measured by various geomagnetic indices such as Ap, Kp, AE, PC or Dst (see Lockwood 2013, and references therein). Herein, we have chosen to use the values of the Dst index (Sugiura 1964) to establish whether the magnetic field perturbations do, in fact, manifest as storms. This is an index that is calculated using four geomagnetic stations situated at low latitudes. Depending on the value of this index, it can be established whether these perturbations are associated with geomagnetic storms or not. In terms of storm intensity, one of the most popular classifications that takes into consideration the minimum value of the Dst index is that of Gonzalez et al. (1994).

The Chapter starts with introductory information about quantitative linguistics notions, like rank--frequency dependence, Zipf's law, frequency spectra, etc. Similarities in distributions of words in texts with level occupation in quantum ensembles hint at a superficial analogy with statistical physics. This enables one to define various parameters for texts based on this physical analogy, including "temperature", "chemical potential", entropy, and some others. Such parameters provide a set of variables to classify texts serving as an example of complex systems. Moreover, texts are perhaps the easiest complex systems to collect and analyze. Similar approaches can be developed to study, for instance, genomes due to well-known linguistic analogies. We consider a couple of approaches to define nucleotide sequences in mitochondrial DNAs and viral RNAs and demonstrate their possible application as an auxiliary tool for comparative analysis of genomes. Finally, we discuss entropy as one of the parameters, which can be easily computed from rank--frequency dependences. Being a discriminating parameter in some problems of classification of complex systems, entropy can be given a proper interpretation only in a limited class of problems. Its overall role and significance remain an open issue so far.