Digital generated image of data. Lemonade is one of this year's hottest IPOs and a key reason for this is the company's heavy investments in AI (Artificial Intelligence). The company has used this technology to develop bots to handle the purchase of policies and the managing of claims. Then how does a company like this create AI models? Well, as should be no surprise, it is complex and susceptible to failure.

Humans ability to transfer knowledge through teaching is one of the essential aspects for human intelligence. A human teacher can track the knowledge of students to customize the teaching on students needs. With the rise of online education platforms, there is a similar need for machines to track the knowledge of students and tailor their learning experience. This is known as the Knowledge Tracing (KT) problem in the literature. Effectively solving the KT problem would unlock the potential of computer-aided education applications such as intelligent tutoring systems, curriculum learning, and learning materials' recommendation. Moreover, from a more general viewpoint, a student may represent any kind of intelligent agents including both human and artificial agents. Thus, the potential of KT can be extended to any machine teaching application scenarios which seek for customizing the learning experience for a student agent (i.e., a machine learning model). In this paper, we provide a comprehensive and systematic review for the KT literature. We cover a broad range of methods starting from the early attempts to the recent state-of-the-art methods using deep learning, while highlighting the theoretical aspects of models and the characteristics of benchmark datasets. Besides these, we shed light on key modelling differences between closely related methods and summarize them in an easy-to-understand format. Finally, we discuss current research gaps in the KT literature and possible future research and application directions.

We consider the general problem of online convex optimization with time-varying additive constraints in the presence of predictions for the next cost and constraint functions. A novel primal-dual algorithm is designed by combining a Follow-The-Regularized-Leader iteration with prediction-adaptive dynamic steps. The algorithm achieves $\mathcal O(T^{\frac{3-\beta}{4}})$ regret and $\mathcal O(T^{\frac{1+\beta}{2}})$ constraint violation bounds that are tunable via parameter $\beta\!\in\![1/2,1)$ and have constant factors that shrink with the predictions quality, achieving eventually $\mathcal O(1)$ regret for perfect predictions. Our work extends the FTRL framework for this constrained OCO setting and outperforms the respective state-of-the-art greedy-based solutions, without imposing conditions on the quality of predictions, the cost functions or the geometry of constraints, beyond convexity.

Federated learning (FL) provides a high efficient decentralized machine learning framework, where the training data remains distributed at remote clients in a network. Though FL enables a privacy-preserving mobile edge computing framework using IoT devices, recent studies have shown that this approach is susceptible to poisoning attacks from the side of remote clients. To address the poisoning attacks on FL, we provide a \textit{two-phase} defense algorithm called {Lo}cal {Ma}licious Facto{r} (LoMar). In phase I, LoMar scores model updates from each remote client by measuring the relative distribution over their neighbors using a kernel density estimation method. In phase II, an optimal threshold is approximated to distinguish malicious and clean updates from a statistical perspective. Comprehensive experiments on four real-world datasets have been conducted, and the experimental results show that our defense strategy can effectively protect the FL system. {Specifically, the defense performance on Amazon dataset under a label-flipping attack indicates that, compared with FG+Krum, LoMar increases the target label testing accuracy from $96.0\%$ to $98.8\%$, and the overall averaged testing accuracy from $90.1\%$ to $97.0\%$.

TEHRAN – The roadmap for the development of artificial intelligence (AI) was drafted after a year of scientific work with the participation of academic and industrial figures from both the public and the private sector. The study of the national artificial intelligence development roadmap document started a year ago at the Research Institute of Information and Communication Technology, which was officially completed by the end of November this year. The document is presented in two general sections, "application development" and "development of enablers", Mohammad-Shahram Moein, head of innovation and development center of artificial intelligence at the Research Institute of Information and Communication Technology, said. Iran is in 13th place among the top countries in artificial intelligence by the total number of publications in 2021."In the application development section, the main goal is to use artificial intelligence in priority areas such as health, transportation, and agriculture, and in the next priorities, areas such as education, industry, and environment have been considered. In the development of enablers, we have considered the training of specialized manpower, the development of infrastructure, and the development of the innovation system," he explained.

Unlike many valuable resources, real-time data is both abundant and growing rapidly. But it also needs to be handled with great care. That was one of the key takeaways from an online workshop produced by Science Business' Data Rules group, which explored what the rapid growth in real-time data means for artificial intelligence (AI). Real-time data is increasingly feeding machine learning systems that then adjust the algorithms they use to make decisions, such as which news item to display on your screen or which product to recommend. "With AI, especially, you want to make sure that the data that you have is consistent, replicable and also valid," noted Chris Atherton, senior research engagement officer at GÉANT, who described how his organisation transmits data captured by the European Space Agency's satellites to researchers across the world.

A few months back I came across a video that started off with a riddle: A boy, who is about to interview in a big company, is in the car with his father. The boy gets a call from the CEO of the company he is about to interview with, and when he answers the call, the CEO says "Good luck son, you've got this". Participants were asked how this was possible?. Do you have any guesses before reading further? Some of them guessed the CEO could be the grandfather, or it could be a pre-recorded call from the father, or the guy has two fathers, and some even guessed that the boy's name was'son'.

We seek to appoint an assistant or associate professor in the general area of Uncertainty in AI, who is passionate about research as well as teaching. We particularly welcome excellent candidates that can contribute to foundational aspects of AI and machine learning.The successful candidate will help with developing and/or delivering courses in DAI cluster, such as Foundations of AI, Explainable AI, Text Mining, Reinforcement Learning, Uncertainty Representation and Reasoning, and Generative Models, and will supervise students at all levels. The working language in the department and across the university is English. An important aspect of TU/e's vision on education is that research and education go hand in hand, both at Bachelor and Master level.Next to your research, education is an important part of your job.The TU/e helps its scientific staff to further develop their teaching skills by offering a training program that leads to an official teaching certification from Dutch Universities (Basic Teaching Qualification).Furthermore, you should have: Are you inspired to work for the exciting Department of Mathematics and Computer Science at TU Eindhoven? We're looking for you as our new faculty member to expand our academic staff in the Data and AI cluster.

Glass Imaging is looking for a Senior Deep Learning / Computer Vision Algorithms Engineer, to work on advanced problems in computational photography. You would be responsible for bringing the latest cutting edge Computer Vision models into production on embedded devices and smartphones, from investigating, developing and training Deep Learning models and algorithms, to optimizing them for high throughput real time imaging applications. Founded by former Apple Engineers who brought you Portrait Mode and other iPhone camera features, Glass is building the future of miniaturized imaging that delivers astonishing image quality and user experience. You'd be joining a unique team of creative and enthusiastic engineers with a passion and track record for revolutionizing the world of photography. We are a funded early-stage startup with great benefits (including stock options, competitive pay and health insurance), a small (but growing) and friendly team.

Deep neural networks (DNNs) have achieved the state of the art performance in numerous fields. However, DNNs need high computation times, and people always expect better performance in a lower computation. Therefore, we study the human somatosensory system and design a neural network (SpinalNet) to achieve higher accuracy with fewer computations. Hidden layers in traditional NNs receive inputs in the previous layer, apply activation function, and then transfer the outcomes to the next layer. In the proposed SpinalNet, each layer is split into three splits: 1) input split, 2) intermediate split, and 3) output split. Input split of each layer receives a part of the inputs. The intermediate split of each layer receives outputs of the intermediate split of the previous layer and outputs of the input split of the current layer. The number of incoming weights becomes significantly lower than traditional DNNs. The SpinalNet can also be used as the fully connected or classification layer of DNN and supports both traditional learning and transfer learning. We observe significant error reductions with lower computational costs in most of the DNNs. Traditional learning on the VGG-5 network with SpinalNet classification layers provided the state-of-the-art (SOTA) performance on QMNIST, Kuzushiji-MNIST, EMNIST (Letters, Digits, and Balanced) datasets. Traditional learning with ImageNet pre-trained initial weights and SpinalNet classification layers provided the SOTA performance on STL-10, Fruits 360, Bird225, and Caltech-101 datasets. The scripts of the proposed SpinalNet are available at the following link: https://github.com/dipuk0506/SpinalNet