Data Systems Laboratory, School of Engineering and Applied Sciences Harvard University, Cambridge, MA USA =============DRAFT July 18, 2023====================== Data science is not a science. It is a research a theory of value that defines the nature, value, paradigm. As data science is in its surpass science - our most powerful research infancy, its axiology can only be speculated. Such paradigm - in enabling knowledge discovery that an axiology can aid in understanding and defining is changing our world[10]. This paper explores and data science and recognizing potenUal benefits, evaluates its remarkable, definiUve features. We present the history and nature of data science and offer Modern data science is in its infancy. Emerging candidate definiUons of essenUal data science slowly since 1962 and rapidly since 2000, data concepts required to discuss its axiology. Within a science is a fundamentally new field of inquiry, decade, this remarkable new research paradigm one of the most acUve, powerful, and rapidly will be seen as a milestone in human knowledge evolving innovaUons of the 21st century. Yet we are just beginning to data science as a Promethean Moment[10] that understand and define it. Due to based on single invenUons, e.g., the prinUng press, its infancy, many definiUons are independent, this moment is based on a meta-technology Essen'al data science concepts data science community to achieve such a Data science (the data science research paradigm) definiUon. To problem solving based on its unique ability to contribute to an iniUal assessment and definiUon computaUonally analyze data to discover insights of data science, this paper proposes an iniUal into moUvaUng domain problems where the axiology of data science. A comprehensive data science axiology is (i.e., learning from data) of data science research A meta technology is used to produce new technology and knowledge hence can be applicable to most human endeavors. Data about, discover, arUculate, and validate the true science results are probabilis5c, correla5onal, nature of the ul5mate ques5ons about natural, possibly fragile or specific to the analysis method observable phenomena as new knowledge about or dataset, cannot be proven complete or correct, those phenomena. ScienUfic results are defini5ve, and lack explana5ons and interpreta5ons for the conclusive, casual, robust, universal knowledge of mo5va5ng domain problem[46]. Like all research paradigms, science and discovery conducted by applying the data science data science are complementary.

Modeling discourse -- the linguistic phenomena that go beyond individual sentences, is a fundamental yet challenging aspect of natural language processing (NLP). However, existing evaluation benchmarks primarily focus on the evaluation of inter-sentence properties and overlook critical discourse phenomena that cross sentences. To bridge the gap, we propose Disco-Bench, a benchmark that can evaluate intra-sentence discourse properties across a diverse set of NLP tasks, covering understanding, translation, and generation. Disco-Bench consists of 9 document-level testsets in the literature domain, which contain rich discourse phenomena (e.g. cohesion and coherence) in Chinese and/or English. For linguistic analysis, we also design a diagnostic test suite that can examine whether the target models learn discourse knowledge. We totally evaluate 20 general-, in-domain and commercial models based on Transformer, advanced pretraining architectures and large language models (LLMs). Our results show (1) the challenge and necessity of our evaluation benchmark; (2) fine-grained pretraining based on literary document-level training data consistently improves the modeling of discourse information. We will release the datasets, pretrained models, and leaderboard, which we hope can significantly facilitate research in this field: https://github.com/longyuewangdcu/Disco-Bench.

Predictive student models are increasingly used in learning environments due to their ability to enhance educational outcomes and support stakeholders in making informed decisions. However, predictive models can be biased and produce unfair outcomes, leading to potential discrimination against some students and possible harmful long-term implications. This has prompted research on fairness metrics meant to capture and quantify such biases. Nonetheless, so far, existing fairness metrics used in education are predictive performance-oriented, focusing on assessing biased outcomes across groups of students, without considering the behaviors of the models nor the severity of the biases in the outcomes. Therefore, we propose a novel metric, the Model Absolute Density Distance (MADD), to analyze models' discriminatory behaviors independently from their predictive performance. We also provide a complementary visualization-based analysis to enable fine-grained human assessment of how the models discriminate between groups of students. We evaluate our approach on the common task of predicting student success in online courses, using several common predictive classification models on an open educational dataset. We also compare our metric to the only predictive performance-oriented fairness metric developed in education, ABROCA. Results on this dataset show that: (1) fair predictive performance does not guarantee fair models' behaviors and thus fair outcomes, (2) there is no direct relationship between data bias and predictive performance bias nor discriminatory behaviors bias, and (3) trained on the same data, models exhibit different discriminatory behaviors, according to different sensitive features too. We thus recommend using the MADD on models that show satisfying predictive performance, to gain a finer-grained understanding on how they behave and to refine models selection and their usage.

We present NusaCrowd, a collaborative initiative to collect and unify existing resources for Indonesian languages, including opening access to previously non-public resources. Through this initiative, we have brought together 137 datasets and 118 standardized data loaders. The quality of the datasets has been assessed manually and automatically, and their value is demonstrated through multiple experiments. NusaCrowd's data collection enables the creation of the first zero-shot benchmarks for natural language understanding and generation in Indonesian and the local languages of Indonesia. Furthermore, NusaCrowd brings the creation of the first multilingual automatic speech recognition benchmark in Indonesian and the local languages of Indonesia. Our work strives to advance natural language processing (NLP) research for languages that are under-represented despite being widely spoken.

Multi-hop machine reading comprehension is a challenging task in natural language processing as it requires more reasoning ability across multiple documents. Spectral models based on graph convolutional networks have shown good inferring abilities and lead to competitive results. However, the analysis and reasoning of some are inconsistent with those of humans. Inspired by the concept of grandmother cells in cognitive neuroscience, we propose a heterogeneous graph attention network model named ClueReader to imitate the grandmother cell concept. The model is designed to assemble the semantic features in multi-level representations and automatically concentrate or alleviate information for reasoning through the attention mechanism. The name ClueReader is a metaphor for the pattern of the model: it regards the subjects of queries as the starting points of clues, takes the reasoning entities as bridge points, considers the latent candidate entities as grandmother cells, and the clues end up in candidate entities. The proposed model enables the visualization of the reasoning graph, making it possible to analyze the importance of edges connecting entities and the selectivity in the mention and candidate nodes, which is easier to comprehend empirically. Evaluations on the open-domain multi-hop reading dataset WikiHop and drug-drug interaction dataset MedHop proved the validity of ClueReader and showed the feasibility of its application of the model in the molecular biology domain.

The use of Machine Learning (ML) has rapidly spread across several fields, having encountered many applications in Structural Dynamics and Vibroacoustic (SD\&V). The increasing capabilities of ML to unveil insights from data, driven by unprecedented data availability, algorithms advances and computational power, enhance decision making, uncertainty handling, patterns recognition and real-time assessments. Three main applications in SD\&V have taken advantage of these benefits. In Structural Health Monitoring, ML detection and prognosis lead to safe operation and optimized maintenance schedules. System identification and control design are leveraged by ML techniques in Active Noise Control and Active Vibration Control. Finally, the so-called ML-based surrogate models provide fast alternatives to costly simulations, enabling robust and optimized product design. Despite the many works in the area, they have not been reviewed and analyzed. Therefore, to keep track and understand this ongoing integration of fields, this paper presents a survey of ML applications in SD\&V analyses, shedding light on the current state of implementation and emerging opportunities. The main methodologies, advantages, limitations, and recommendations based on scientific knowledge were identified for each of the three applications. Moreover, the paper considers the role of Digital Twins and Physics Guided ML to overcome current challenges and power future research progress. As a result, the survey provides a broad overview of the present landscape of ML applied in SD\&V and guides the reader to an advanced understanding of progress and prospects in the field.

Adam is a commonly used stochastic optimization algorithm in machine learning. However, its convergence is still not fully understood, especially in the non-convex setting. This paper focuses on exploring hyperparameter settings for the convergence of vanilla Adam and tackling the challenges of non-ergodic convergence related to practical application. The primary contributions are summarized as follows: firstly, we introduce precise definitions of ergodic and non-ergodic convergence, which cover nearly all forms of convergence for stochastic optimization algorithms. Meanwhile, we emphasize the superiority of non-ergodic convergence over ergodic convergence. Secondly, we establish a weaker sufficient condition for the ergodic convergence guarantee of Adam, allowing a more relaxed choice of hyperparameters. On this basis, we achieve the almost sure ergodic convergence rate of Adam, which is arbitrarily close to $o(1/\sqrt{K})$. More importantly, we prove, for the first time, that the last iterate of Adam converges to a stationary point for non-convex objectives. Finally, we obtain the non-ergodic convergence rate of $O(1/K)$ for function values under the Polyak-Lojasiewicz (PL) condition. These findings build a solid theoretical foundation for Adam to solve non-convex stochastic optimization problems.

Author Note We have no known conflict of interest to disclose. Correspondence concerning this article should be addressed to Hyemin Han, University of Alabama, Box 872031, Tuscaloosa, AL 35487, United States. Email: hyemin.han@ua.edu 2 Potential Benefits of Employing Large Language Models in Research in Moral Education and Development Abstract Recently, computer scientists have developed large language models (LLMs) by training prediction models with large-scale language corpora and human reinforcements. The LLMs have become one promising way to implement artificial intelligence with accuracy in various fields. Interestingly, recent LLMs possess emergent functional features that emulate sophisticated human cognition, especially in-context learning and the chain of thought, which were unavailable in previous prediction models. In this paper, I will examine how LLMs might contribute to moral education and development research. To achieve this goal, I will review the most recently published conference papers and ArXiv preprints to overview the novel functional features implemented in LLMs. I also intend to conduct brief experiments with ChatGPT to investigate how LLMs behave while addressing ethical dilemmas and external feedback. The results suggest that LLMs might be capable of solving dilemmas based on reasoning and revising their reasoning process with external input. Furthermore, a preliminary experimental result from the moral exemplar test may demonstrate that exemplary stories can elicit moral elevation in LLMs as do they among human participants. I will discuss the potential implications of LLMs on research on moral education and development with the results. Keywords: Large language models, Artificial intelligence, Moral reasoning, Moral exemplar, Simulation 3 Introduction One of the most impactful recent developments in computer science is large language models (LLMs) (Grossmann et al., 2023), which implement advanced artificial intelligence.

We investigate the mathematical capabilities of two iterations of ChatGPT (released 9-January-2023 and 30-January-2023) and of GPT-4 by testing them on publicly available datasets, as well as hand-crafted ones, using a novel methodology. In contrast to formal mathematics, where large databases of formal proofs are available (e.g., the Lean Mathematical Library), current datasets of natural-language mathematics, used to benchmark language models, either cover only elementary mathematics or are very small. We address this by publicly releasing two new datasets: GHOSTS and miniGHOSTS. These are the first natural-language datasets curated by working researchers in mathematics that (1) aim to cover graduate-level mathematics, (2) provide a holistic overview of the mathematical capabilities of language models, and (3) distinguish multiple dimensions of mathematical reasoning. These datasets also test whether ChatGPT and GPT-4 can be helpful assistants to professional mathematicians by emulating use cases that arise in the daily professional activities of mathematicians. We benchmark the models on a range of fine-grained performance metrics. For advanced mathematics, this is the most detailed evaluation effort to date. We find that ChatGPT can be used most successfully as a mathematical assistant for querying facts, acting as a mathematical search engine and knowledge base interface. GPT-4 can additionally be used for undergraduate-level mathematics but fails on graduate-level difficulty. Contrary to many positive reports in the media about GPT-4 and ChatGPT's exam-solving abilities (a potential case of selection bias), their overall mathematical performance is well below the level of a graduate student. Hence, if your goal is to use ChatGPT to pass a graduate-level math exam, you would be better off copying from your average peer!

Spiking neural networks (SNNs) are bio-plausible computing models with high energy efficiency. The temporal dynamics of neurons and synapses enable them to detect temporal patterns and generate sequences. While Backpropagation Through Time (BPTT) is traditionally used to train SNNs, it is not suitable for online learning of embedded applications due to its high computation and memory cost as well as extended latency. Previous works have proposed online learning algorithms, but they often utilize highly simplified spiking neuron models without synaptic dynamics and reset feedback, resulting in subpar performance. In this work, we present Spatiotemporal Online Learning for Synaptic Adaptation (SOLSA), specifically designed for online learning of SNNs composed of Leaky Integrate and Fire (LIF) neurons with exponentially decayed synapses and soft reset. The algorithm not only learns the synaptic weight but also adapts the temporal filters associated to the synapses. Compared to the BPTT algorithm, SOLSA has much lower memory requirement and achieves a more balanced temporal workload distribution. Moreover, SOLSA incorporates enhancement techniques such as scheduled weight update, early stop training and adaptive synapse filter, which speed up the convergence and enhance the learning performance. When compared to other non-BPTT based SNN learning, SOLSA demonstrates an average learning accuracy improvement of 14.2%. Furthermore, compared to BPTT, SOLSA achieves a 5% higher average learning accuracy with a 72% reduction in memory cost.