Large language models have recently attracted significant attention due to their impressive performance on a variety of tasks. ChatGPT developed by OpenAI is one such implementation of a large, pre-trained language model that has gained immense popularity among early adopters, where certain users go to the extent of characterizing it as a disruptive technology in many domains. Understanding such early adopters' sentiments is important because it can provide insights into the potential success or failure of the technology, as well as its strengths and weaknesses. In this paper, we conduct a mixed-method study using 10,732 tweets from early ChatGPT users. We first use topic modelling to identify the main topics and then perform an in-depth qualitative sentiment analysis of each topic. Our results show that the majority of the early adopters have expressed overwhelmingly positive sentiments related to topics such as Disruptions to software development, Entertainment and exercising creativity. Only a limited percentage of users expressed concerns about issues such as the potential for misuse of ChatGPT, especially regarding topics such as Impact on educational aspects. We discuss these findings by providing specific examples for each topic and then detail implications related to addressing these concerns for both researchers and users.

The problem of answering questions using knowledge from pre-trained language models (LMs) and knowledge graphs (KGs) presents two challenges: given a QA context (question and answer choice), methods need to (i) identify relevant knowledge from large KGs, and (ii) perform joint reasoning over the QA context and KG. In this work, we propose a new model, QA-GNN, which addresses the above challenges through two key innovations: (i) relevance scoring, where we use LMs to estimate the importance of KG nodes relative to the given QA context, and (ii) joint reasoning, where we connect the QA context and KG to form a joint graph, and mutually update their representations through graph neural networks. We evaluate our model on QA benchmarks in the commonsense (CommonsenseQA, OpenBookQA) and biomedical (MedQA-USMLE) domains. QA-GNN outperforms existing LM and LM+KG models, and exhibits capabilities to perform interpretable and structured reasoning, e.g., correctly handling negation in questions.

In this paper, we model the reward of Autoregressive processes naturally arise in a a sequential decision-making problem as an AR process large variety of real-world scenarios, including where its parameters depend on the action selected by the e.g., stock markets, sell forecasting, weather agent at every round. This scenario can be regarded as an prediction, advertising, and pricing. When extension of the multi-armed bandit (MAB, Lattimore & addressing a sequential decision-making problem Szepesvári, 2020) problem, in which an AR process governs in such a context, the temporal dependence the temporal structure of the observed rewards that between consecutive observations should be is, through the action-dependent AR parameters, that are properly accounted for converge to the optimal unknown to the agent. It is worth mentioning that such decision policy. In this work, we propose a novel a scenario displays notable differences compared to more online learning setting, named Autoregressive traditional non-stationary MABs (Gur et al., 2014). Indeed, Bandits (ARBs), in which the observed reward in the presented scenario, we can exploit the knowledge that follows an autoregressive process of order k, the underlying process is AR and, more importantly, that whose parameters depend on the action the such a dynamic depends on the agent's action.

Knowledge Distillation (KD) has been extensively used for natural language understanding (NLU) tasks to improve a small model's (a student) generalization by transferring the knowledge from a larger model (a teacher). Although KD methods achieve state-of-the-art performance in numerous settings, they suffer from several problems limiting their performance. It is shown in the literature that the capacity gap between the teacher and the student networks can make KD ineffective. Additionally, existing KD techniques do not mitigate the noise in the teacher's output: modeling the noisy behaviour of the teacher can distract the student from learning more useful features. We propose a new KD method that addresses these problems and facilitates the training compared to previous techniques. Inspired by continuation optimization, we design a training procedure that optimizes the highly non-convex KD objective by starting with the smoothed version of this objective and making it more complex as the training proceeds. Our method (Continuation-KD) achieves state-of-the-art performance across various compact architectures on NLU (GLUE benchmark) and computer vision tasks (CIFAR-10 and CIFAR-100).

Causal inference is the process of using assumptions, study designs, and estimation strategies to draw conclusions about the causal relationships between variables based on data. This allows researchers to better understand the underlying mechanisms at work in complex systems and make more informed decisions. In many settings, we may not fully observe all the confounders that affect both the treatment and outcome variables, complicating the estimation of causal effects. To address this problem, a growing literature in both causal inference and machine learning proposes to use Instrumental Variables (IV). This paper serves as the first effort to systematically and comprehensively introduce and discuss the IV methods and their applications in both causal inference and machine learning. First, we provide the formal definition of IVs and discuss the identification problem of IV regression methods under different assumptions. Second, we categorize the existing work on IV methods into three streams according to the focus on the proposed methods, including two-stage least squares with IVs, control function with IVs, and evaluation of IVs. For each stream, we present both the classical causal inference methods, and recent developments in the machine learning literature. Then, we introduce a variety of applications of IV methods in real-world scenarios and provide a summary of the available datasets and algorithms. Finally, we summarize the literature, discuss the open problems and suggest promising future research directions for IV methods and their applications. We also develop a toolkit of IVs methods reviewed in this survey at https://github.com/causal-machine-learning-lab/mliv.

Event cameras are novel bio-inspired sensors that measure per-pixel brightness differences asynchronously. Recovering brightness from events is appealing since the reconstructed images inherit the high dynamic range (HDR) and high-speed properties of events; hence they can be used in many robotic vision applications and to generate slow-motion HDR videos. However, state-of-the-art methods tackle this problem by training an event-to-image Recurrent Neural Network (RNN), which lacks explainability and is difficult to tune. In this work we show, for the first time, how tackling the combined problem of motion and brightness estimation leads us to formulate event-based image reconstruction as a linear inverse problem that can be solved without training an image reconstruction RNN. Instead, classical and learning-based regularizers are used to solve the problem and remove artifacts from the reconstructed images. The experiments show that the proposed approach generates images with visual quality on par with state-of-the-art methods despite only using data from a short time interval. State-of-the-art results are achieved using an image denoising Convolutional Neural Network (CNN) as the regularization function. The proposed regularized formulation and solvers have a unifying character because they can be applied also to reconstruct brightness from the second derivative. Additionally, the formulation is attractive because it can be naturally combined with super-resolution, motion-segmentation and color demosaicing. Code is available at https://github.com/tub-rip/event_based_image_rec_inverse_problem

Such skills can be learnt through methods such as imitation learning [1] or reinforce ment learning [2]. However, since these methods are data-driven, they would require training across a huge range of environments if the task is to execute the skills whilst avoiding collisions with the environment. And since this is typically impractical, these skills are typically trained in open, unconstrained environments, without enabling transfer of the skill to more complex, cluttered environments. This then raises the following question: how do we maintain the high performance of a skill that was trained in obstacle-free environments, whilst ensuring that it does not collide with obstacles in new cluttered environments? In complex environments, reactively avoiding obstacles by adding small control adjustments to a learned skill might not be enough, resulting in a need to completely alter the skill and re-learn how to complete the task in the new environment.

In recent years, deep learning (DL) models have demonstrated remarkable achievements on non-trivial tasks such as speech recognition and natural language understanding. One of the significant contributors to its success is the proliferation of end devices that acted as a catalyst to provide data for data-hungry DL models. However, computing DL training and inference is the main challenge. Usually, central cloud servers are used for the computation, but it opens up other significant challenges, such as high latency, increased communication costs, and privacy concerns. To mitigate these drawbacks, considerable efforts have been made to push the processing of DL models to edge servers. Moreover, the confluence point of DL and edge has given rise to edge intelligence (EI). This survey paper focuses primarily on the fifth level of EI, called all in-edge level, where DL training and inference (deployment) are performed solely by edge servers. All in-edge is suitable when the end devices have low computing resources, e.g., Internet-of-Things, and other requirements such as latency and communication cost are important in mission-critical applications, e.g., health care. Firstly, this paper presents all in-edge computing architectures, including centralized, decentralized, and distributed. Secondly, this paper presents enabling technologies, such as model parallelism and split learning, which facilitate DL training and deployment at edge servers. Thirdly, model adaptation techniques based on model compression and conditional computation are described because the standard cloud-based DL deployment cannot be directly applied to all in-edge due to its limited computational resources. Fourthly, this paper discusses eleven key performance metrics to evaluate the performance of DL at all in-edge efficiently. Finally, several open research challenges in the area of all in-edge are presented.

More than three years ago, this editor sat down with Sam Altman for a small event in San Francisco soon after he'd left his role as the president of Y Combinator to become CEO of the AI company he co-founded in 2015 with Elon Musk and others, OpenAI. At the time, Altman described OpenAI's potential in language that sounded outlandish to some. Altman said, for example, that the opportunity with artificial general intelligence -- machine intelligence that can solve problems as well as a human -- is so great that if OpenAI managed to crack it, the outfit could "maybe capture the light cone of all future value in the universe." He said that the company was "going to have to not release research" because it was so powerful. Asked if OpenAI was guilty of fear-mongering -- Musk has repeatedly called all organizations developing AI to be regulated -- Altman talked about the dangers of not thinking about "societal consequences" when "you're building something on an exponential curve."

For the past year, the Chinese government has been conducting some of the earliest experiments in building regulatory tools to govern artificial intelligence (AI). In that process, China is trying to tackle a problem that will soon face governments around the world: Can regulators gain meaningful insight into the functioning of algorithms, and ensure they perform within acceptable bounds? One particular tool deserves attention both for its impact within China, and for the lessons technologists and policymakers in other countries can draw from it: a mandatory registration system created by China's internet regulator for recommendation algorithms. Although the full details of the registry are not public, by digging into its online instruction manual, we can reveal new insights into China's emerging regulatory architecture for algorithms. The algorithm registry was created by China's 2022 regulation on recommendation algorithms (English translation), which came into effect in March of this year and was led by the Cyberspace Administration of China (CAC).