Artificial intelligence (AI) researchers have been developing and refining large language models (LLMs) that exhibit remarkable capabilities across a variety of domains and tasks, challenging our understanding of learning and cognition. The latest model developed by OpenAI, GPT-4, was trained using an unprecedented scale of compute and data. In this paper, we report on our investigation of an early version of GPT-4, when it was still in active development by OpenAI. We contend that (this early version of) GPT-4 is part of a new cohort of LLMs (along with ChatGPT and Google's PaLM for example) that exhibit more general intelligence than previous AI models. We discuss the rising capabilities and implications of these models. We demonstrate that, beyond its mastery of language, GPT-4 can solve novel and difficult tasks that span mathematics, coding, vision, medicine, law, psychology and more, without needing any special prompting. Moreover, in all of these tasks, GPT-4's performance is strikingly close to human-level performance, and often vastly surpasses prior models such as ChatGPT. Given the breadth and depth of GPT-4's capabilities, we believe that it could reasonably be viewed as an early (yet still incomplete) version of an artificial general intelligence (AGI) system. In our exploration of GPT-4, we put special emphasis on discovering its limitations, and we discuss the challenges ahead for advancing towards deeper and more comprehensive versions of AGI, including the possible need for pursuing a new paradigm that moves beyond next-word prediction. We conclude with reflections on societal influences of the recent technological leap and future research directions.

ChatGPT is another large language model (LLM) inline but due to its performance and ability to converse effectively, it has gained a huge popularity amongst research as well as industrial community. Recently, many studies have been published to show the effectiveness, efficiency, integration, and sentiments of chatGPT and other LLMs. In contrast, this study focuses on the important aspects that are mostly overlooked, i.e. sustainability, privacy, digital divide, and ethics and suggests that not only chatGPT but every subsequent entry in the category of conversational bots should undergo Sustainability, PrivAcy, Digital divide, and Ethics (SPADE) evaluation. This paper discusses in detail about the issues and concerns raised over chatGPT in line with aforementioned characteristics. We support our hypothesis by some preliminary data collection and visualizations along with hypothesized facts. We also suggest mitigations and recommendations for each of the concerns. Furthermore, we also suggest some policies and recommendations for AI policy act, if designed by the governments.

Individual human decision-makers may benefit from different forms of support to improve decision outcomes. However, a key question is which form of support will lead to accurate decisions at a low cost. In this work, we propose learning a decision support policy that, for a given input, chooses which form of support, if any, to provide. We consider decision-makers for whom we have no prior information and formalize learning their respective policies as a multi-objective optimization problem that trades off accuracy and cost. Using techniques from stochastic contextual bandits, we propose $\texttt{THREAD}$, an online algorithm to personalize a decision support policy for each decision-maker, and devise a hyper-parameter tuning strategy to identify a cost-performance trade-off using simulated human behavior. We provide computational experiments to demonstrate the benefits of $\texttt{THREAD}$ compared to offline baselines. We then introduce $\texttt{Modiste}$, an interactive tool that provides $\texttt{THREAD}$ with an interface. We conduct human subject experiments to show how $\texttt{Modiste}$ learns policies personalized to each decision-maker and discuss the nuances of learning decision support policies online for real users.

Research on automated essay scoring has become increasing important because it serves as a method for evaluating students' written-responses at scale. Scalable methods for scoring written responses are needed as students migrate to online learning environments resulting in the need to evaluate large numbers of written-response assessments. The purpose of this study is to describe and evaluate three active learning methods than can be used to minimize the number of essays that must be scored by human raters while still providing the data needed to train a modern automated essay scoring system. The three active learning methods are the uncertainty-based, the topological-based, and the hybrid method. These three methods were used to select essays included as part of the Automated Student Assessment Prize competition that were then classified using a scoring model that was training with the bidirectional encoder representations from transformer language model. All three active learning methods produced strong results, with the topological-based method producing the most efficient classification. Growth rate accuracy was also evaluated. The active learning methods produced different levels of efficiency under different sample size allocations but, overall, all three methods were highly efficient and produced classifications that were similar to one another.

We consider the infinite-horizon LQR control problem. Motivated by competitive analysis in online learning, as a criterion for controller design we introduce the dynamic regret, defined as the difference between the LQR cost of a causal controller (that has only access to past disturbances) and the LQR cost of the \emph{unique} clairvoyant one (that has also access to future disturbances) that is known to dominate all other controllers. The regret itself is a function of the disturbances, and we propose to find a causal controller that minimizes the worst-case regret over all bounded energy disturbances. The resulting controller has the interpretation of guaranteeing the smallest regret compared to the best non-causal controller that can see the future. We derive explicit formulas for the optimal regret and for the regret-optimal controller for the state-space setting. These explicit solutions are obtained by showing that the regret-optimal control problem can be reduced to a Nehari extension problem that can be solved explicitly. The regret-optimal controller is shown to be linear and can be expressed as the sum of the classical $H_2$ state-feedback law and an $n$-th order controller ($n$ is the state dimension), and its construction simply requires a solution to the standard LQR Riccati equation and two Lyapunov equations. Simulations over a range of plants demonstrate that the regret-optimal controller interpolates nicely between the $H_2$ and the $H_\infty$ optimal controllers, and generally has $H_2$ and $H_\infty$ costs that are simultaneously close to their optimal values. The regret-optimal controller thus presents itself as a viable option for control systems design.

In a randomized exam, each student is asked a small number of random questions from a large question bank. The predominant grading rule is simple averaging, i.e., calculating grades by averaging scores on the questions each student is asked, which is fair ex-ante, over the randomized questions, but not fair ex-post, on the realized questions. The fair grading problem is to estimate the average grade of each student on the full question bank. The maximum-likelihood estimator for the Bradley-Terry-Luce model on the bipartite student-question graph is shown to be consistent with high probability when the number of questions asked to each student is at least the cubed-logarithm of the number of students. In an empirical study on exam data and in simulations, our algorithm based on the maximum-likelihood estimator significantly outperforms simple averaging in prediction accuracy and ex-post fairness even with a small class and exam size.

As a result of transformation processes, the German labor market is highly dependent on vocational training, retraining and continuing education. To match training seekers and offers, we present a novel approach towards the automated detection of access to education and training in German training offers and advertisements. We will in particular focus on (a) general school and education degrees and schoolleaving certificates, (b) professional experience, (c) a previous apprenticeship and (d) a list of skills provided by the German Federal Employment Agency. This novel approach combines several methods: First, we provide a mapping of synonyms in education combining different qualifications and adding deprecated terms. Second, we provide a rule-based matching to identify the need for professional experience or apprenticeship. However, not all access requirements can be matched due to incompatible data schemata or non-standardizes requirements, e.g initial tests or interviews. While we can identify several shortcomings, the presented approach offers promising results for two data sets: training and re-training advertisements.

Riiid, a leading provider of AI-powered education solutions and member company of Born2Global Centre, has announced the launch of R.test, an AI-powered Platform designed to help students prepare for standardized tests, including the digital SAT and ACT. With R.test, students can predict test scores with high accuracy in only a quarter of the time it takes to complete full mock tests. R.test's AI engine provides students with an overview of their current test-taking habits and predicts the correctness of answers for questions that they have not solved. The platform analyzes students' weaknesses and offers actionable guidance on how to improve, including an AI-curated selection of relevant practice questions. It also provides students with an analysis of their time spent answering questions, highlighting areas where they need to work more quickly.

If you've reached adulthood and wished you spent a little more time trying to learn an instrument as a kid, you're not alone. There are more than one million people using Skoove's interactive piano and keyboard lessons to learn the piano today. Skoove is one of the best ways to learn the piano, employing an award-winning system to help you learn and practice notes, chords, and techniques by playing your favorite songs. The AI-enhanced app listens and adapts to your playing, giving you individual feedback to focus on your weaknesses, and planning a curriculum for you. With 400 lessons and thousands of instructional videos, you'll get a tailored training regimen that will let you learn in the way that makes the most sense for you.

There are all sorts of options for learning modern natural language processing, notably those options with neural networks techniques. For example, there is freely-available course material from world class universities such as Stanford's Natural Language Processing with Deep Learning, among others. There are also courses, paid and otherwise, from independent non-university sources such as Coursera and fast.ai. There is a wide variety of quality books that have been published over the recent few years which are topical and up-to-date. Today, if you want to learn modern natural language processing techniques, there is no excuse for not doing so.