Machine learning enables the development of new, supplemental, and empowering tools that can either expand existing technologies or invent new ones. In education, space exists for a tool that supports generic student course review formats to organize and recapitulate students' views on the pedagogical practices to which they are exposed. Often, student opinions are gathered with a general comment section that solicits their feelings towards their courses without polling specifics about course contents. Herein, we show a novel approach to summarizing and organizing students' opinions via analyzing their sentiment towards a course as a function of the language/vocabulary used to convey their opinions about a class and its contents. This analysis is derived from their responses to a general comment section encountered at the end of post-course review surveys. This analysis, accomplished with Python, LaTeX, and Google's Natural Language API, allows for the conversion of unstructured text data into both general and topic-specific sub-reports that convey students' views in a unique, novel way.

Machine learning models work better when curated features are provided to them. Feature engineering methods have been usually used as a preprocessing step to obtain or build a proper feature set. In late years, autoencoders (a specific type of symmetrical neural network) have been widely used to perform representation learning, proving their competitiveness against classical feature engineering algorithms. The main obstacle in the use of autoencoders is finding a good architecture, a process that most experts confront manually. An automated autoencoder architecture search procedure, based on evolutionary methods, is proposed in this paper. The methodology is tested against nine heterogeneous data sets. The obtained results show the ability of this approach to find better architectures, able to concentrate most of the useful information in a minimized coding, in a reduced time.

There are a plethora of methods and algorithms that solve the classical multi-label document classification. However, when it comes to deployment and usage in an industry setting, most, if not all the contemporary approaches fail to address some of the vital aspects or requirements of an ideal solution: i. ability to operate on variable-length texts and rambling documents. ii. catastrophic forgetting problem. iii. modularity when it comes to online learning and updating the model. iv. ability to spotlight relevant text while producing the prediction, i.e. visualizing the predictions. v. ability to operate on imbalanced or skewed datasets. vi. scalability. The paper describes the significance of these problems in detail and proposes a unique neural network architecture that addresses the above problems. The proposed architecture views documents as a sequence of sentences and leverages sentence-level embeddings for input representation. A hydranet-like architecture is designed to have granular control over and improve the modularity, coupled with a weighted loss driving task-specific heads. In particular, two specific mechanisms are compared: Bi-LSTM and Transformer-based. The architecture is benchmarked on some of the popular benchmarking datasets such as Web of Science - 5763, Web of Science - 11967, BBC Sports, and BBC News datasets. The experimental results reveal that the proposed model outperforms the existing methods by a substantial margin. The ablation study includes comparisons of the impact of the attention mechanism and the application of weighted loss functions to train the task-specific heads in the hydranet.

The goal of sequential event prediction is to estimate the next event based on a sequence of historical events, with applications to sequential recommendation, user behavior analysis and clinical treatment. In practice, the next-event prediction models are trained with sequential data collected at one time and need to generalize to newly arrived sequences in remote future, which requires models to handle temporal distribution shift from training to testing. In this paper, we first take a data-generating perspective to reveal a negative result that existing approaches with maximum likelihood estimation would fail for distribution shift due to the latent context confounder, i.e., the common cause for the historical events and the next event. Then we devise a new learning objective based on backdoor adjustment and further harness variational inference to make it tractable for sequence learning problems. On top of that, we propose a framework with hierarchical branching structures for learning context-specific representations. Comprehensive experiments on diverse tasks (e.g., sequential recommendation) demonstrate the effectiveness, applicability and scalability of our method with various off-the-shelf models as backbones.

This paper introduces corpus-guided top-down synthesis as a mechanism for synthesizing library functions that capture common functionality from a corpus of programs in a domain specific language (DSL). The algorithm builds abstractions directly from initial DSL primitives, using syntactic pattern matching of intermediate abstractions to intelligently prune the search space and guide the algorithm towards abstractions that maximally capture shared structures in the corpus. We present an implementation of the approach in a tool called Stitch and evaluate it against the state-of-the-art deductive library learning algorithm from DreamCoder. Our evaluation shows that Stitch is 3-4 orders of magnitude faster and uses 2 orders of magnitude less memory while maintaining comparable or better library quality (as measured by compressivity). We also demonstrate Stitch's scalability on corpora containing hundreds of complex programs that are intractable with prior deductive approaches and show empirically that it is robust to terminating the search procedure early -- further allowing it to scale to challenging datasets by means of early stopping.

We study a new paradigm of knowledge transfer that aims at encoding graph topological information into graph neural networks (GNNs) by distilling knowledge from a teacher GNN model trained on a complete graph to a student GNN model operating on a smaller or sparser graph. To this end, we revisit the connection between thermodynamics and the behavior of GNN, based on which we propose Neural Heat Kernel (NHK) to encapsulate the geometric property of the underlying manifold concerning the architecture of GNNs. A fundamental and principled solution is derived by aligning NHKs on teacher and student models, dubbed as Geometric Knowledge Distillation. We develop non- and parametric instantiations and demonstrate their efficacy in various experimental settings for knowledge distillation regarding different types of privileged topological information and teacher-student schemes.

Recent advances in coreset methods have shown that a selection of representative datapoints can replace massive volumes of data for Bayesian inference, preserving the relevant statistical information and significantly accelerating subsequent downstream tasks. Existing variational coreset constructions rely on either selecting subsets of the observed datapoints, or jointly performing approximate inference and optimizing pseudodata in the observed space akin to inducing points methods in Gaussian Processes. So far, both approaches are limited by complexities in evaluating their objectives for general purpose models, and require generating samples from a typically intractable posterior over the coreset throughout inference and testing. In this work, we present a black-box variational inference framework for coresets that overcomes these constraints and enables principled application of variational coresets to intractable models, such as Bayesian neural networks. We apply our techniques to supervised learning problems, and compare them with existing approaches in the literature for data summarization and inference.

Distribution shift occurs when the test distribution differs from the training distribution, and it can considerably degrade performance of machine learning models deployed in the real world. Temporal shifts -- distribution shifts arising from the passage of time -- often occur gradually and have the additional structure of timestamp metadata. By leveraging timestamp metadata, models can potentially learn from trends in past distribution shifts and extrapolate into the future. While recent works have studied distribution shifts, temporal shifts remain underexplored. To address this gap, we curate Wild-Time, a benchmark of 5 datasets that reflect temporal distribution shifts arising in a variety of real-world applications, including patient prognosis and news classification. On these datasets, we systematically benchmark 13 prior approaches, including methods in domain generalization, continual learning, self-supervised learning, and ensemble learning. We use two evaluation strategies: evaluation with a fixed time split (Eval-Fix) and evaluation with a data stream (Eval-Stream). Eval-Fix, our primary evaluation strategy, aims to provide a simple evaluation protocol, while Eval-Stream is more realistic for certain real-world applications. Under both evaluation strategies, we observe an average performance drop of 20% from in-distribution to out-of-distribution data. Existing methods are unable to close this gap. Code is available at https://wild-time.github.io/.

Earlier this month, New York City public schools blocked access to the popular artificial intelligence tool ChatGPT. Educators are concerned that students could use this technology to write papers – the tool wasn't even a month old when a college professor in South Carolina caught a student using it to write an essay in philosophy class. Darren Hick of Furman University joins John Yang to discuss.

The education community has been abuzz with the rise of ChatGPT, an artificial intelligence tool that can write anything with just a simple prompt. Most of the conversation has been centered on the extent to which students will use the chat bot--but ChatGPT could also fundamentally change the nature of teachers' jobs. So far, teachers have used--or considered using--the chat bot to plan lessons, put together rubrics, offer students feedback on assignments, respond to parent emails, and write letters of recommendation, among other tasks. While some educators worry about the implications of automating these parts of teaching, others say that the tool can save them hours of work, freeing up time for student interactions or their personal life. After all, a typical teacher works about 54 hours a week, but just under half of that time is devoted to directly teaching students, according to a nationally representative survey of teachers conducted by the EdWeek Research Center last year. Just under a third of teachers said if they could spend less time on any one task, it would be general administrative work.