Imbalanced Data Robust Online Continual Learning Based on Evolving Class Aware Memory Selection and Built-In Contrastive Representation Learning

We consider robust online learning in time-variant neural network regression models. Using a state space representation for the neural network's weight evolution in time we derive weight estimates by maximizing posterior modes via the Fisher scoring algorithm. By taking the family of densities as the output error cost function we get a robust error measure suitable for handling additive outliers. Fisher scoring was implemented using a forward backward pass of fixed length through the data set for every time step resulting in so-called online smoothing algorithms. Furthermore, we derive an EM-type algorithm for approximate maximum likelihood estimation of unknown hyperparameters. Our experiments show that online posterior mode weight smoothing outperforms standard online methods like online backpropagation and extended Kalman filtering, both for Gaussian measurements and non-Gaussian measurements with additive outliers.

Online, distance, and eLearning (ODeL) continue to gain recognition as a mandatory component of delivery of education in institutions of higher learning (IHL) around the world following the outbreak of coronavirus disease (COVID-19). This paradigm shift is informed by the need to ensure uninterrupted, valuable, and safe learning experiences for learners during the pandemic. However, governments ordered the closure of schools and colleges following the declaration of COVID-19 as a world pandemic by the World Health Organization (WHO). A report by United Nations Educational, Scientific and Cultural Organization revealed that there was a significant loss of schooling time following the closure of educational facilities which affected over 1.5 billion learners in 194 nations globally. This study explored the use of online approaches to intensify online learning efficacy in IHL. Data collection was conducted using qualitative methods and data analysis done using themes and sub-themes. Findings from this study indicate that students’ engagements on discussion forums are consistent with collaborative learning. Results further support the view that regular, prompt, and meaningful feedback is critical in promoting constructive learning and reflection among students. Based on the findings of this study, practical implications are discussed for stakeholders interested in establishing and strengthening effective delivery of online learning content to enhance students’ learning experiences.

The purpose of this mixed method parallel/convergent research was to ascertain educators' perspectives of and responses to COVID 19 in the Jamaican education system and assess its state of readiness for online teaching and learning. It further sought to provide critical insights on the lessons learned in crisis management and steps required to propel Jamaica into a future of robust online teaching and learning. The findings revealed that while most educators owned their own devices, had internet connectivity, and could satisfactorily navigate the various online platforms, there were issues with the level and scope of training and support they received. Further, the major drawback was the low number of students that were able to access the online space. This undeniably indicates greater need for effective leadership and management especially in times of crises. So, the major recommendations were for continuous professional development in crisis management and other areas as well as resource support to be offered so that the most vulnerable students can benefit equitably.

Learning analytics (LA) is the "measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs" (Siemens & Gasevic, 2012).Originally, "analytic" refers to a way of using data to support decision-making and understanding a domain.Essential LA components are (1) data, (2) goals or (research) questions, optionally based on educational theory, (3) measures that give information about goal attainment or (research) construct, optionally (4) descriptive or predictive models that use these values as variables, and (5)

This paper introduces an intelligent system developed for the online learning and adaptive control of the human glucose metabolism. The measurement process of the glucose concentration usually involves significant time-delay. The metabolic system was modelled by a 4/sup th/ order digital model with time-delay, and a three term microcontroller is employed in real-time processing of such system. The adaptive control of this system was achieved using a recently developed criterion such that the closed-loop dynamics closely track a described and well-behaved trajectory. The instrumental variable estimator (IV) was used for online learning of such system by successively adjusting the model parameters based on a set of input-output patterns incorporating an output disturbance of stochastic nature. The design and implementation of this intelligent system is presented by simulating a realistic system using a LabTop computer.

We present an unsupervised speaker identification system for personal annotations of conversations and meetings. The system dynamically learns new speakers and recognizes already known speakers using one audio channel and speech-independent modeling. Multiple personal systems could collaborate in robust unsupervised speaker identification and online learning. The system was optimized for real-time operation on a DSP system that can be worn during daily activities. The system was evaluated on the freely available 24-speaker Augmented Multiparty Interaction dataset. For 5 s recognition time, the system achieves 81% recognition rate. Collaboration between four identification systems resulted in a performance increase of up to 17%, however even two collaborating systems yield an performance improvement. A prototypical wearable DSP implementation could continuously operate for more than 8 hours from a 4.1 Ah battery.

The COVID-19 pandemic and its restrictions increased the adoption of online learning even in low-income countries. The adoption of online teaching methods may have affected teaching and learning, particularly in settings where it was used for the first time. This study was conducted to explore the perceptions of medical and nursing students regarding the impact of online delivery of problem-based learning (PBL) on students learning and academic performance during COVID-19 imposed restrictions.Methods and materialsThis was a qualitative study among fourth and fifth-year nursing and medical undergraduate students at Busitema University Faculty of Health Sciences. Four focused group discussions were conducted and the interviews focused on students’ perceptions, experiences, and attitudes toward the PBL process conducted online and its likely impact on their learning. Braun and Clarke’s thematic analysis was used for qualitative data analysis.ResultsFour themes were identified that represented perceptions of online PBL on learning: transition to online learning; perceived benefits of online learning; limited learning and poor performance; and lost soft and practical skills. During the initial stages of introduction to online PBL learning, students transitioning to online had to adapt and familiarize themselves with online learning following the introduction of online learning. Students perceived that learning was less online compared to face-to-face sessions because of reduced learner engagement, concentration, motivation, peer-to-peer learning, and limited opportunities for practical sessions. Online learning was thought to increase students’ workload in the form of a number of assessments which was thought to reduce learning. Online tutorials were perceived to reduce the acquisition of soft skills like confidence, communication, leadership, and practical or clinical skills. While learning was thought to be less during online teaching, it was noted to allow continued learning during the lockdown, to be flexible, enhance self-drive and opportunity for work, solve infrastructure problems, and protect them from COVID-19 infectionConclusionGenerally, online learning enabled continuity and flexibility of learning. However, online PBL learning was perceived to be less engaging compared to traditional classroom-based PBL. Online PBL was seen to deter students from acquiring critical generic and clinical skills inherently found in traditional PBL. Innovative pedagogical measures should be adopted to avoid reduced learning noted in the online teaching methods to ensure the successful adoption of online teaching and learning in the post-COVID-19 era.

Background. The relevance of the presented topic should be associated with the existential situation characterized by the active development of the information society and its transition to a knowledge society, in which the values ​​associated with self-education and self-development of the individual prevail. Therefore, the strategy of continuous learning «throughout life» is in demand, allowing to satisfy the needs of the individual for self-development and self-improvement. The subject of this study is the concept of continuous lifelong learning, which acts as a practice of «self-care». Purpose. The main goal of the article is to explicate the axiological potential of continuous lifelong learning and to show that such learning acts primarily as «self-care», allowing the individual to develop and improve. Materials and methods. For the logic of development and solution of research problems, such methods as conceptual modeling, interpretation, and the method of contextual analysis were of great importance. The main provisions of the article contribute to the formation of a general conceptual idea of ​​the potential of continuous lifelong learning both for self-development and improvement of the individual, and for understanding the existence of the individual in the knowledge society. Results. The article raises the question: how or with the help of what can the strategy of continuous lifelong learning be implemented? It is stated that the demand for continuous lifelong learning of a large number of people can be satisfied only with the help of e-learning and distance learning technologies. And online courses, in particular massive open online courses (MOOCs), are the most popular and promising. It is argued that thanks to online courses, each person is potentially capable of creating a space for (self) education, and in a situation of personal readiness for continuous learning, MOOCs as a special case of online courses appear to be a phenomenon that contributes to the implementation of the practice of «self-care».

The complex and changeable inland river scenes resulting out of frequent occlusions of ships in the available tracking methods are not accurate enough to estimate the motion state of the target ship leading to object tracking drift or even loss. In view of this, an attempt is made to propose a robust online learning ship tracking algorithm based on the Siamese network and the region proposal network. Firstly, the algorithm combines the off-line Siamese network classification score and the online classifier score for discriminative learning, and establishes an occlusion determination mechanism according to the classification the fusion score. When the target is in the occlusion state, the target template is not updated, and the global search mechanism is activated to relocate the target, thereby avoiding object tracking drift. Secondly, an efficient adaptive online update strategy, UpdateNet, is introduced to improve the template degradation in the tracking process. Finally, on comparing the state-of-the-art tracking algorithms on the inland river ship datasets, the experimental results of the proposed algorithm show strong robustness in occlusion scenarios with an accuracy and success rate of 56.8% and 57.2% respectively. Supportive source codes for this research are publicly available at https://github.com/Libra-jing/SiamOL.

Profound Learning (PL) can result from online interactions that support practices for deep, lifelong learning. Distance educators can initiate, facilitate, and maintain PL by encouraging thoughtful rather than superficial learning using Universal Design for Learning (UDL). The inclusive nature of UDL provides multiple mechanisms to find that deeper meaning. In this article, concepts and practices drawn from Profound Learning Theory are integrated into UDL guidelines and connected to distance learning to support the development of deep, meaningful, and robust online learning.

Malicious data manipulation reduces the effectiveness of machine learning techniques, which rely on accurate knowledge of the input data. Motivated by real-world applications in network flow classification, we address the problem of robust online learning with delayed feedback in the presence of malicious data generators that attempt to gain favorable classification outcome by manipulating the data features. When the feedback delay is static, we propose online algorithms termed ROLC-NC and ROLC-C when the malicious data generators are non-clairvoyant and clairvoyant, respectively. We then consider the dynamic delay case, for which we propose online algorithms termed ROLC-NC-D and ROLC-C-D when the malicious data generators are non-clairvoyant and clairvoyant, respectively. We derive regret bounds for these four algorithms and show that they are sub-linear under mild conditions. We further evaluate the proposed algorithms in network flow classification via extensive experiments using real-world data traces. Our experimental results demonstrate that the proposed algorithms can approach the performance of an optimal static offline classifier that is not under attack, while outperforming the same offline classifier when tested with a mixture of normal and manipulated data.

Neuromorphic Very Large Scale Integration (VLSI) hardware offers a low-power and compact electronic substrate for implementing distributed plastic synapses for artificial systems.However, the technology used for constructing analog neuromorphic circuits has limited resolution and high intrinsic variability, leading to large circuit mismatch.Consequently, neuromorphic synapse

Malicious data manipulation reduces the effectiveness of machine learning techniques, which rely on accurate knowledge of the input data. Motivated by real-world applications in network flow classification, we address the problem of robust online learning with delayed feedback in the presence of malicious data generators that attempt to gain favorable classification outcome by manipulating the data features. We propose online algorithms termed ROLC-NC and ROLC-C when the malicious data generators are non-clairvoyant and clairvoyant, respectively. We derive regret bounds for both algorithms and show that they are sub-linear under mild conditions. We further evaluate the proposed algorithms in network flow classification via extensive experiments using real-world data traces. Our experimental results demonstrate that both algorithms can approach the performance of an optimal static offline classifier that is not under attack, while outperforming the same offline classifier when tested with a mixture of normal and manipulated data.

Robots often need to learn the human's reward function online, during the current interaction. This real-time learning requires fast but approximate learning rules: when the human's behavior is noisy or suboptimal, current approximations can result in unstable robot learning. Accordingly, in this paper we seek to enhance the robustness and convergence properties of gradient descent learning rules when inferring the human's reward parameters. We model the robot's learning algorithm as a dynamical system over the human preference parameters, where the human's true (but unknown) preferences are the equilibrium point. This enables us to perform Lyapunov stability analysis to derive the conditions under which the robot's learning dynamics converge. Our proposed algorithm (StROL) uses these conditions to learn robust-by-design learning rules: given the original learning dynamics, StROL outputs a modified learning rule that now converges to the human's true parameters under a larger set of human inputs. In practice, these autonomously generated learning rules can correctly infer what the human is trying to convey, even when the human is noisy, biased, and suboptimal. Across simulations and a user study we find that StROL results in a more accurate estimate and less regret than state-of-the-art approaches for online reward learning. See videos and code here: https://github.com/VT-Collab/StROL_RAL

Handbook of Robust Low-Rank and Sparse Matrix Decomposition: Applications in Image and Video Processing shows you how robust subspace learning and tracking by decomposition into low-rank and sparse matrices provide a suitable framework for computer vision applications. Incorporating both existing and new ideas, the book conveniently gives you one-stop access to a number of different decompositions, algorithms, implementations, and benchmarking techniques. Divided into five parts, the book begins with an overall introduction to robust principal component analysis (PCA) via decomposition into low-rank and sparse matrices. The second part addresses robust matrix factorization/completion problems while the third part focuses on robust online subspace estimation, learning, and tracking. Covering applications in image and video processing, the fourth part discusses image analysis, image denoising, motion saliency detection, video coding, key frame extraction, and hyperspectral video processing. The final part presents resources and applications in background/foreground separation for video surveillance. With contributions from leading teams around the world, this handbook provides a complete overview of the concepts, theories, algorithms, and applications related to robust low-rank and sparse matrix decompositions. It is designed for researchers, developers, and graduate students in computer vision, image and video processing, real-time architecture, machine learning, and data mining.