Wray Buntine

VinUniversity · College of Engineering and Computer Science

Adversarial robustness, domain generalization and dataset biases are three active lines of research contributing to out-of-distribution (OOD) evaluation on neural NLP models. However, a comprehensive, integrated discussion of the three research lines is still lacking in the literature. This survey will 1) compare the three lines of research under a...

Adversarial robustness, domain generalization and dataset biases are three active lines of research contributing to out-of-distribution (OOD) evaluation on neural NLP models. However, a comprehensive, integrated discussion of the three research lines is still lacking in the literature. In this survey, we 1) compare the three lines of research under...

Dialogue acts (DAs) can represent conversational actions of tutors or students that take place during tutoring dialogues. Automating the identification of DAs in tutoring dialogues is significant to the design of dialogue-based intelligent tutoring systems. Many prior studies employ machine learning models to classify DAs in tutoring dialogues and...

Dialogue Acts (DAs) can be used to explain what expert tutors do and what students know during the tutoring process. Most empirical studies adopt the random sampling method to obtain sentence samples for manual annotation of DAs, which are then used to train DA classifiers. However, these studies have paid little attention to sample informativeness...

Current work in named entity recognition (NER) uses either cross entropy (CE) or conditional random fields (CRF) as the objective/loss functions to optimize the underlying NER model. Both of these traditional objective functions for the NER problem generally produce adequate performance when the data distribution is balanced and there are sufficien...

Cross-domain graph anomaly detection (CD-GAD) describes the problem of detecting anomalous nodes in an unlabelled target graph using auxiliary, related source graphs with labelled anomalous and normal nodes. Although it presents a promising approach to address the notoriously high false positive issue in anomaly detection, little work has been done...

Large language models (LLMs) have shown great abilities of solving various natural language tasks in different domains. Due to the training objective of LLMs and their pretraining data, LLMs are not very well equipped for tasks involving structured data generation. We propose a framework, Prompting with Iterative Verification (PiVe), to improve gra...

Unlabelled: Recently, research on short text topic models has addressed the challenges of social media datasets. These models are typically evaluated using automated measures. However, recent work suggests that these evaluation measures do not inform whether the topics produced can yield meaningful insights for those examining social media data. E...

Dialogue acts (DAs) can represent conversational actions of tutors or students that take place during tutoring dialogues. Automating the identification of DAs in tutoring dialogues is significant to the design of dialogue-based intelligent tutoring systems. Many prior studies employ machine learning models to classify DAs in tutoring dialogues and...

Dialogue Acts (DAs) can be used to explain what expert tutors do and what students know during the tutoring process. Most empirical studies adopt the random sampling method to obtain sentence samples for manual annotation of DAs, which are then used to train DA classifiers. However, these studies have paid little attention to sample informativeness...

A critical challenge in developing successful AI application projects is the lack of understanding of business characteristics and user experiences. In many cases, AI systems developed by AI experts would not be effectively utilised by the staff and are rejected by customers. The high risk of failure makes most managers hesitant in adopting AI for...

Automating the classification of instructional strategies from a large-scale online tutorial dialogue corpus is indispensable to the design of dialogue-based intelligent tutoring systems. Despite many existing studies employing supervised machine learning (ML) models to automate the classification process, they concluded that building a well-perfor...

Current work in named entity recognition (NER) uses either cross entropy (CE) or conditional random fields (CRF) as the objective/loss functions to optimize the underlying NER model. Both of these traditional objective functions for the NER problem generally produce adequate performance when the data distribution is balanced and there are sufficien...

Cross-domain graph anomaly detection (CD-GAD) describes the problem of detecting anomalous nodes in an unlabelled target graph using auxiliary, related source graphs with labelled anomalous and normal nodes. Although it presents a promising approach to address the notoriously high false positive issue in anomaly detection, little work has been done...

Hierarchical stochastic processes, such as the hierarchical Dirichlet process, hold an important position as a modelling tool in statistical machine learning, and are even used in deep neural networks. They allow, for instance, networks of probability vectors to be used in general statistical modelling, intrinsically supporting information sharing...

Domain adaptation is an effective solution to data scarcity in low-resource scenarios. However, when applied to token-level tasks such as bioNER, domain adaptation methods often suffer from the challenging linguistic characteristics that clinical narratives possess, which leads to unsatisfactory performance. In this paper, we present a simple yet e...

Text simplification is the task of rewriting a text so that it is readable and easily understood. In this paper, we propose a simple yet novel unsupervised sentence simplification system that harnesses parsing structures together with sentence embeddings to produce linguistically effective simplifications. This means our model is capable of introdu...

Recent unsupervised GNN based graph anomaly detection (GAD) methods adopt specific mechanisms designed for anomaly detection. This is in contrast to earlier methods that utilise components such as graph autoencoders that were designed for more general use-cases. However, these newer methods only lead to a modest increase in detection accuracy at th...

Modern deep learning methods constitute incredibly powerful tools to tackle a myriad of challenging problems. However, since deep learning methods operate as black boxes, the uncertainty associated with their predictions is often challenging to quantify. Bayesian statistics offer a formalism to understand and quantify the uncertainty associated wit...

We study acquisition functions for active learning (AL) for text classification. The Expected Loss Reduction (ELR) method focuses on a Bayesian estimate of the reduction in classification error, recently updated with Mean Objective Cost of Uncertainty (MOCU). We convert the ELR framework to estimate the increase in (strictly proper) scores like log...

Background Secondary use of electronic health record's (EHR) data requires evaluation of data quality (DQ) for fitness of use. While multiple frameworks exist for quantifying DQ, there are no guidelines for the evaluation of DQ failures identified through such frameworks. Objectives This study proposes a systematic approach to evaluate DQ failures...

Deep generative models have been widely used in several areas of NLP, and various techniques have been proposed to augment them or address their training challenges. In this paper, we propose a simple modification to Variational Autoencoders (VAEs) by using an Isotropic Gaussian Posterior (IGP) that allows for better utilisation of their latent rep...

This paper proposes a transformer over transformer framework, called Transformer$^2$, to perform neural text segmentation. It consists of two components: bottom-level sentence encoders using pre-trained transformers, and an upper-level transformer-based segmentation model based on the sentence embeddings. The bottom-level component transfers the pr...

Neural topic models (NTMs) apply deep neural networks to topic modelling. Despite their success, NTMs generally ignore two important aspects: (1) only document-level word count information is utilized for the training, while more fine-grained sentence-level information is ignored, and (2) external semantic knowledge regarding documents, sentences a...

Multilingual Neural Machine Translation (MNMT) trains a single NMT model that supports translation between multiple languages, rather than training separate models for different languages. Learning a single model can enhance the low-resource translation by leveraging data from multiple languages. However, the performance of an MNMT model is highly...

Topic modelling has been a successful technique for text analysis for almost twenty years. When topic modelling met deep neural networks, there emerged a new and increasingly popular research area, neural topic models, with nearly a hundred models developed and a wide range of applications in neural language understanding such as text generation, s...

Recently, listwise collaborative filtering (CF) algorithms are attracting increasing interest due to their efficiency and prediction quality. Different from rating-oriented (pointwise) CF, they recommend a preference ranking of items to each user without estimating the absolute value of the ratings. In practice, there are extensive side information...

Purpose Full text of a document is a rich source of information that can be used to provide meaningful topics. The purpose of this paper is to demonstrate how to use citation context (CC) in the full text to identify the cited topics and citing topics efficiently and effectively by employing automatic text analysis algorithms. Design/methodology/a...

Your article is protected by copyright and all rights are held exclusively by Springer Science+Business Media, LLC, part of Springer Nature. This e-offprint is for personal use only and shall not be self-archived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own...

Pseudo-labeling is a key component in semi-supervised learning (SSL). It relies on iteratively using the model to generate artificial labels for the unlabeled data to train against. A common property among its various methods is that they only rely on the model's prediction to make labeling decisions without considering any prior knowledge about th...

Software Quality Assurance (SQA) planning aims to define proactive plans, such as defining maximum file size, to prevent the occurrence of software defects in future releases. To aid this, defect prediction models have been proposed to generate insights as the most important factors that are associated with software quality. Such insights that are...

Topic modelling has been a successful technique for text analysis for almost twenty years. When topic modelling met deep neural networks, there emerged a new and increasingly popular research area, neural topic models, with over a hundred models developed and a wide range of applications in neural language understanding such as text generation, sum...

Software Quality Assurance (SQA) planning aims to define proactive plans, such as defining maximum file size, to prevent the occurrence of software defects in future releases. To aid this, defect prediction models have been proposed to generate insights as the most important factors that are associated with software quality. Such insights that are...

Predicting (1) when the next hospital admission occurs and (2) what will happen in the next admission about a patient by mining electronic health record (EHR) data can provide granular readmission predictions to assist clinical decision making. Recurrent neural network (RNN) and point process models are usually employed in modelling temporal sequen...

Supervised learning, characterized by both discriminative and generative learning, seeks to predict the values of single (or sometimes multiple) predefined target attributes based on a predefined set of predictor attributes. For applications where the information available and predictions to be made may vary from instance to instance, we propose th...

Scarcity of parallel sentence-pairs poses a significant hurdle for training high-quality Neural Machine Translation (NMT) models in bilingually low-resource scenarios. A standard approach is transfer learning, which involves taking a model trained on a high-resource language-pair and fine-tuning it on the data of the low-resource MT condition of in...

Bayesian network classifiers are, functionally, an interesting class of models, because they can be learnt out-of-core, i.e. without needing to hold the whole training data in main memory. The selective K-dependence Bayesian network classifier (SKDB) is state of the art in this class of models and has shown to rival random forest (RF) on problems w...

In this paper, we present a new topic modelling approach via the theory of optimal transport (OT). Specifically, we present a document with two distributions: a distribution over the words (doc-word distribution) and a distribution over the topics (doc-topic distribution). For one document, the doc-word distribution is the observed, sparse, low-lev...

Modern deep learning methods have equipped researchers and engineers with incredibly powerful tools to tackle problems that previously seemed impossible. However, since deep learning methods operate as black boxes, the uncertainty associated with their predictions is often challenging to quantify. Bayesian statistics offer a formalism to understand...

BACKGROUND In the absence of a vaccine or curative treatment, non-pharmaceutical interventions regimes have been implemented by governments around the world, to slow the spread of COVID-19. The success of these NPIs has varied between countries and is likely to relate to the degree of uptake and adherence by the community. Understanding public atti...

Background Nonpharmaceutical interventions (NPIs) (such as wearing masks and social distancing) have been implemented by governments around the world to slow the spread of COVID-19. To promote public adherence to these regimes, governments need to understand the public perceptions and attitudes toward NPI regimes and the factors that influence them...

As we rely more and more on machine learning models for real-life decision-making, being able to understand and trust the predictions becomes ever more important. Local explainer models have recently been introduced to explain the predictions of complex machine learning models at the instance level. In this paper, we propose Local Rule-based Model...

Graph embedding methods are useful for a wide range of graph analysis tasks including link prediction and node classification. Most graph embedding methods learn only the topological structure of graphs. Nevertheless, it has been shown that the incorporation of node attributes is beneficial in improving the expressive power of node embeddings. Howe...

Decision trees are still seeing use in online, non-stationary and embedded contexts, as well as for interpretability. For applications like ranking and cost-sensitive classification, probability estimation trees (PETs) are used. These are built using smoothing or calibration techniques. Older smoothing techniques used counts local to a leaf node, b...

Decision trees are still seeing use in online, non-stationary and embedded contexts, as well as for interpretability. For applications like ranking and cost-sensitive classification, probability estimation trees (PETs) are used. These are built using smoothing or calibration techniques. Older smoothing techniques used counts local to a leaf node, b...

Graph embedding methods transform high-dimensional and complex graph contents into low-dimensional representations. They are useful for a wide range of graph analysis tasks including link prediction, node classification, recommendation and visualization. Most existing approaches represent graph nodes as point vectors in a low-dimensional embedding...

Electronic medical record (EMR) data contains historical sequences of visits of patients, and each visit contains rich information, such as patient demographics, hospital utilisation and medical codes, including diagnosis, procedure and medication codes. Most existing EMR embedding methods capture visit-code associations by constructing input visit...

Graph embedding methods transform high-dimensional and complex graph contents into low-dimensional representations. They are useful for a wide range of graph analysis tasks including link prediction, node classification, recommendation and visualization. Most existing approaches represent graph nodes as point vectors in a low-dimensional embedding...

Besides the text content, documents usually come with rich sets of meta-information, such as categories of documents and semantic/syntactic features of words, like those encoded in word embeddings. Incorporating such meta-information directly into the generative process of topic models can improve modelling accuracy and topic quality, especially in...

As we rely more and more on machine learning models for real-life decision-making, being able to understand and trust the predictions becomes ever more important. Local explainer models have recently been introduced to explain the predictions of complex machine learning models at the instance level. In this paper, we propose Local Rule-based Model...

Many applications, such as text modelling, high-throughput sequencing, and recommender systems, require analysing sparse, high-dimensional, and overdispersed discrete (count/binary) data. With the ability of handling high-dimensional and sparse discrete data, models based on probabilistic matrix factorisation and latent factor analysis have enjoyed...

Computing the probability of unseen documents is a natural evaluation task in topic modeling. Previous work has addressed this problem for the well-known Latent Dirichlet Allocation (LDA) model. However, the same problem for a more general class of topic models, referred here to as Gamma-Poisson Factor Analysis (GaP-FA), remains unexplored, which h...

Recently, considerable research effort has been devoted to developing deep architectures for topic models to learn topic structures. Although several deep models have been proposed to learn better topic proportions of documents, how to leverage the benefits of deep structures for learning word distributions of topics has not yet been rigorously stu...

This paper introduces a novel parameter estimation method for the probability tables of Bayesian Network Classifiers (BNCs), using Hierarchical Dirichlet Processes (HDPs). The main result of this paper is to show that proper parameter estimation allows BNCs to outperform leading learning methods such as Random Forest for both 0-1 loss and RMSE, alb...

A rich variety of models are now in use for unsupervised modelling of text documents, and, in particular, a rich variety of graphical models exist, with and without latent variables. To date, there is inadequate understanding about the comparative performance of these, partly because they are subtly different, and they have been proposed and evalua...

The questions in a crowdsourcing task typically exhibit varying degrees of difficulty and subjectivity. Their joint effects give rise to the variation in responses to the same question by different crowd-workers. This variation is low when the question is easy to answer and objective, and high when it is difficult and subjective. Unfortunately, cur...

Recent advances have demonstrated substantial benefits from learning with both generative and discriminative parameters. On the one hand, generative approaches address the estimation of the parameters of the joint distribution—\(\mathrm{P}(y,\mathbf{x})\), which for most network types is very computationally efficient (a notable exception to this a...

Besides the text content, documents and their associated words usually come with rich sets of meta informa- tion, such as categories of documents and semantic/syntactic features of words, like those encoded in word embeddings. Incorporating such meta information directly into the generative process of topic models can improve modelling accuracy and...

Relational data are usually highly incomplete in practice, which inspires us to leverage side information to improve the performance of community detection and link prediction. This paper presents a Bayesian probabilistic approach that incorporates various kinds of node attributes encoded in binary form in relational models with Poisson likelihood....