Current evaluations of Continual Learning (CL) methods typically assume that there is no constraint on training time and computation. This is an unrealistic assumption for any real-world setting, which motivates us to propose: a practical real-time evaluation of continual learning, in which the stream does not wait for the model to complete training before revealing the next data for predictions. To do this, we evaluate current CL methods with respect to their computational costs. We conduct extensive experiments on CLOC, a large-scale dataset containing 39 million time-stamped images with geolocation labels. We show that a simple baseline outperforms state-of-the-art CL methods under this evaluation, questioning the applicability of existing methods in realistic settings. In addition, we explore various CL components commonly used in the literature, including memory sampling strategies and regularization approaches. We find that all considered methods fail to be competitive against our simple baseline. This surprisingly suggests that the majority of existing CL literature is tailored to a specific class of streams that is not practical. We hope that the evaluation we provide will be the first step towards a paradigm shift to consider the computational cost in the development of online continual learning methods.

ABSTRACT One of the most important magnetic spin structure is the topologically stabilised skyrmion quasi-particle. Its interesting physical properties make them candidates for memory and efficient neuromorphic computation schemes. For the device operation, detection of the position, shape, and size of skyrmions is required and magnetic imaging is typically employed. A frequently used technique is magneto-optical Kerr microscopy where depending on the sample's material composition, temperature, material growing procedures, etc., the measurements suffer from noise, low-contrast, intensity gradients, or other optical artifacts. Conventional image analysis packages require manual treatment, and a more automatic solution is required. We report a convolutional neural network specifically designed for segmentation problems to detect the position and shape of skyrmions in our measurements. The network is tuned using selected techniques to optimize predictions and in particular the number of detected classes is found to govern the performance. The results of this study shows that a well-trained network is a viable method of automating data pre-processing in magnetic microscopy. The approach is easily extendable to other spin structures and other magnetic imaging methods. INTRODUCTION In the last decade, magnetic quasiparticles [1], [2] have raised the interest of researchers due to their interesting properties and their potential applicability in next-generation memory and neuromorphic devices [2] In particular, skyrmions in magnetic thin-films are under investigation due to their topological stabilization, their small size, and their ability to be easily manipulated by spin currents [4], [5].

Recent work in natural language processing (NLP) has yielded appealing results from scaling model parameters and training data; however, using only scale to improve performance means that resource consumption also grows. Such resources include data, time, storage, or energy, all of which are naturally limited and unevenly distributed. This motivates research into efficient methods that require fewer resources to achieve similar results. This survey synthesizes and relates current methods and findings in efficient NLP. We aim to provide both guidance for conducting NLP under limited resources, and point towards promising research directions for developing more efficient methods.

Open-domain dialogue systems aim to interact with humans through natural language texts in an open-ended fashion. Despite the recent success of super large dialogue systems such as ChatGPT, using medium-to-small-sized dialogue systems remains the common practice as they are more lightweight and accessible; however, generating diverse dialogue responses is challenging, especially with smaller models. In this work, we propose an Equal-size Hard Expectation--Maximization (EqHard-EM) algorithm to train a multi-decoder model for diverse dialogue generation. Our algorithm assigns a sample to a decoder in a hard manner and additionally imposes an equal-assignment constraint to ensure that all decoders are well-trained. We provide detailed theoretical analysis to justify our approach. Further, experiments on two large-scale open-domain dialogue datasets verify that our EqHard-EM algorithm generates high-quality diverse responses.

Deep neural networks obtained by standard training have been constantly plagued by adversarial examples. Although adversarial training demonstrates its capability to defend against adversarial examples, unfortunately, it leads to an inevitable drop in the natural generalization. To address the issue, we decouple the natural generalization and the robust generalization from joint training and formulate different training strategies for each one. Specifically, instead of minimizing a global loss on the expectation over these two generalization errors, we propose a bi-expert framework called \emph{Generalist} where we simultaneously train base learners with task-aware strategies so that they can specialize in their own fields. The parameters of base learners are collected and combined to form a global learner at intervals during the training process. The global learner is then distributed to the base learners as initialized parameters for continued training. Theoretically, we prove that the risks of Generalist will get lower once the base learners are well trained. Extensive experiments verify the applicability of Generalist to achieve high accuracy on natural examples while maintaining considerable robustness to adversarial ones. Code is available at https://github.com/PKU-ML/Generalist.

Quantum machine learning is a fast-emerging field that aims to tackle machine learning using quantum algorithms and quantum computing. Due to the lack of physical qubits and an effective means to map real-world data from Euclidean space to Hilbert space, most of these methods focus on quantum analogies or process simulations rather than devising concrete architectures based on qubits. In this paper, we propose a novel hybrid quantum-classical algorithm for graph-structured data, which we refer to as the Ego-graph based Quantum Graph Neural Network (egoQGNN). egoQGNN implements the GNN theoretical framework using the tensor product and unity matrix representation, which greatly reduces the number of model parameters required. When controlled by a classical computer, egoQGNN can accommodate arbitrarily sized graphs by processing ego-graphs from the input graph using a modestly-sized quantum device. The architecture is based on a novel mapping from real-world data to Hilbert space. This mapping maintains the distance relations present in the data and reduces information loss. Experimental results show that the proposed method outperforms competitive state-of-the-art models with only 1.68\% parameters compared to those models.

We demonstrate that, through appropriate prompting, GPT-3 family of models can be triggered to perform iterative behaviours necessary to execute (rather than just write or recall) programs that involve loops, including several popular algorithms found in computer science curricula or software developer interviews. We trigger execution and description of Iterations by Regimenting Self-Attention (IRSA) in one (or a combination) of three ways: 1) Using strong repetitive structure in an example of an execution path of a target program for one particular input, 2) Prompting with fragments of execution paths, and 3) Explicitly forbidding (skipping) self-attention to parts of the generated text. On a dynamic program execution, IRSA leads to larger accuracy gains than replacing the model with the much more powerful GPT-4. IRSA has promising applications in education, as the prompts and responses resemble student assignments in data structures and algorithms classes. Our findings hold implications for evaluating LLMs, which typically target the in-context learning: We show that prompts that may not even cover one full task example can trigger algorithmic behaviour, allowing solving problems previously thought of as hard for LLMs, such as logical puzzles. Consequently, prompt design plays an even more critical role in LLM performance than previously recognized.

There is considerable evidence that machine learning algorithms have better predictive abilities than humans in various financial settings. But, the literature has not tested whether these algorithmic predictions are more rational than human predictions. We study the predictions of corporate earnings from several algorithms, notably linear regressions and a popular algorithm called Gradient Boosted Regression Trees (GBRT). On average, GBRT outperformed both linear regressions and human stock analysts, but it still overreacted to news and did not satisfy rational expectation as normally defined. By reducing the learning rate, the magnitude of overreaction can be minimized, but it comes with the cost of poorer out-of-sample prediction accuracy. Human stock analysts who have been trained in machine learning methods overreact less than traditionally trained analysts. Additionally, stock analyst predictions reflect information not otherwise available to machine algorithms.

The growing prominence of large language models, such as GPT-4 and ChatGPT, has led to increased concerns over academic integrity due to the potential for machine-generated content and paraphrasing. Although studies have explored the detection of human- and machine-paraphrased content, the comparison between these types of content remains underexplored. In this paper, we conduct a comprehensive analysis of various datasets commonly employed for paraphrase detection tasks and evaluate an array of detection methods. Our findings highlight the strengths and limitations of different detection methods in terms of performance on individual datasets, revealing a lack of suitable machine-generated datasets that can be aligned with human expectations. Our main finding is that human-authored paraphrases exceed machine-generated ones in terms of difficulty, diversity, and similarity implying that automatically generated texts are not yet on par with human-level performance. Transformers emerged as the most effective method across datasets with TF-IDF excelling on semantically diverse corpora. Additionally, we identify four datasets as the most diverse and challenging for paraphrase detection.

Oversight is rightly recognised as vital within high-stakes public sector AI applications, where decisions can have profound individual and collective impacts. Much current thinking regarding forms of oversight mechanisms for AI within the public sector revolves around the idea of human decision makers being 'in-the-loop' and thus being able to intervene to prevent errors and potential harm. However, in a number of high-stakes public sector contexts, operational oversight of decisions is made by expert teams rather than individuals. The ways in which deployed AI systems can be integrated into these existing operational team oversight processes has yet to attract much attention. We address this gap by exploring the impacts of AI upon pre-existing oversight of clinical decision-making through institutional analysis. We find that existing oversight is nested within professional training requirements and relies heavily upon explanation and questioning to elicit vital information. Professional bodies and liability mechanisms also act as additional levers of oversight. These dimensions of oversight are impacted, and potentially reconfigured, by AI systems. We therefore suggest a broader lens of 'team-in-the-loop' to conceptualise the system-level analysis required for adoption of AI within high-stakes public sector deployment.