arXiv subjects

Byron C. Wallace

Publications and source records attributed to Byron C. Wallace.

At least 91 records · Page 5Linked to original sources

arXiv2020

ERASER: A Benchmark to Evaluate Rationalized NLP Models

State-of-the-art models in NLP are now predominantly based on deep neural networks that are opaque in terms of how they come to make predictions. This limitation has increased interest in designing more interpretable deep models for NLP that reveal the `reasoning' behind model outputs. But work in this direction has been conducted on different datasets and tasks with correspondingly unique aims and metrics; this makes it difficult to track progress. We propose the Evaluating Rationales And Simple English Reasoning (ERASER) benchmark to advance research on interpretable models in NLP. This benchmark comprises multiple datasets and tasks for which human annotations of "rationales" (supporting evidence) have been collected. We propose several metrics that aim to capture how well the rationales provided by models align with human rationales, and also how faithful these rationales are (i.e., the degree to which provided rationales influenced the corresponding predictions). Our hope is that releasing this benchmark facilitates progress on designing more interpretable NLP systems. The benchmark, code, and documentation are available at https://www.eraserbenchmark.com/

cs.CL↗

arXiv2019

Practical Obstacles to Deploying Active Learning

Active learning (AL) is a widely-used training strategy for maximizing predictive performance subject to a fixed annotation budget. In AL one iteratively selects training examples for annotation, often those for which the current model is most uncertain (by some measure). The hope is that active sampling leads to better performance than would be achieved under independent and identically distributed (i.i.d.) random samples. While AL has shown promise in retrospective evaluations, these studies often ignore practical obstacles to its use. In this paper we show that while AL may provide benefits when used with specific models and for particular domains, the benefits of current approaches do not generalize reliably across models and tasks. This is problematic because in practice one does not have the opportunity to explore and compare alternative AL strategies. Moreover, AL couples the training dataset with the model used to guide its acquisition. We find that subsequently training a successor model with an actively-acquired dataset does not consistently outperform training on i.i.d. sampled data. Our findings raise the question of whether the downsides inherent to AL are worth the modest and inconsistent performance gains it tends to afford.

cs.LG↗

arXiv2019

Learning to Identify Patients at Risk of Uncontrolled Hypertension Using Electronic Health Records Data

Hypertension is a major risk factor for stroke, cardiovascular disease, and end-stage renal disease, and its prevalence is expected to rise dramatically. Effective hypertension management is thus critical. A particular priority is decreasing the incidence of uncontrolled hypertension. Early identification of patients at risk for uncontrolled hypertension would allow targeted use of personalized, proactive treatments. We develop machine learning models (logistic regression and recurrent neural networks) to stratify patients with respect to the risk of exhibiting uncontrolled hypertension within the coming three-month period. We trained and tested models using EHR data from 14,407 and 3,009 patients, respectively. The best model achieved an AUROC of 0.719, outperforming the simple, competitive baseline of relying prediction based on the last BP measure alone (0.634). Perhaps surprisingly, recurrent neural networks did not outperform a simple logistic regression for this task, suggesting that linear models should be included as strong baselines for predictive tasks using EHR

cs.CY↗

arXiv2019

Predicting Annotation Difficulty to Improve Task Routing and Model Performance for Biomedical Information Extraction

Modern NLP systems require high-quality annotated data. In specialized domains, expert annotations may be prohibitively expensive. An alternative is to rely on crowdsourcing to reduce costs at the risk of introducing noise. In this paper we demonstrate that directly modeling instance difficulty can be used to improve model performance, and to route instances to appropriate annotators. Our difficulty prediction model combines two learned representations: a `universal' encoder trained on out-of-domain data, and a task-specific encoder. Experiments on a complex biomedical information extraction task using expert and lay annotators show that: (i) simply excluding from the training data instances predicted to be difficult yields a small boost in performance; (ii) using difficulty scores to weight instances during training provides further, consistent gains; (iii) assigning instances predicted to be difficult to domain experts is an effective strategy for task routing. Our experiments confirm the expectation that for specialized tasks expert annotations are higher quality than crowd labels, and hence preferable to obtain if practical. Moreover, augmenting small amounts of expert data with a larger set of lay annotations leads to further improvements in model performance.

cs.CL↗

arXiv2019

Attention is not Explanation

Attention mechanisms have seen wide adoption in neural NLP models. In addition to improving predictive performance, these are often touted as affording transparency: models equipped with attention provide a distribution over attended-to input units, and this is often presented (at least implicitly) as communicating the relative importance of inputs. However, it is unclear what relationship exists between attention weights and model outputs. In this work, we perform extensive experiments across a variety of NLP tasks that aim to assess the degree to which attention weights provide meaningful `explanations' for predictions. We find that they largely do not. For example, learned attention weights are frequently uncorrelated with gradient-based measures of feature importance, and one can identify very different attention distributions that nonetheless yield equivalent predictions. Our findings show that standard attention modules do not provide meaningful explanations and should not be treated as though they do. Code for all experiments is available at https://github.com/successar/AttentionExplanation.

cs.CL↗

arXiv2019

An Analysis of Attention over Clinical Notes for Predictive Tasks

The shift to electronic medical records (EMRs) has engendered research into machine learning and natural language technologies to analyze patient records, and to predict from these clinical outcomes of interest. Two observations motivate our aims here. First, unstructured notes contained within EMR often contain key information, and hence should be exploited by models. Second, while strong predictive performance is important, interpretability of models is perhaps equally so for applications in this domain. Together, these points suggest that neural models for EMR may benefit from incorporation of attention over notes, which one may hope will both yield performance gains and afford transparency in predictions. In this work we perform experiments to explore this question using two EMR corpora and four different predictive tasks, that: (i) inclusion of attention mechanisms is critical for neural encoder modules that operate over notes fields in order to yield competitive performance, but, (ii) unfortunately, while these boost predictive performance, it is decidedly less clear whether they provide meaningful support for predictions.

cs.CL↗

arXiv2019

Inferring Which Medical Treatments Work from Reports of Clinical Trials

How do we know if a particular medical treatment actually works? Ideally one would consult all available evidence from relevant clinical trials. Unfortunately, such results are primarily disseminated in natural language scientific articles, imposing substantial burden on those trying to make sense of them. In this paper, we present a new task and corpus for making this unstructured evidence actionable. The task entails inferring reported findings from a full-text article describing a randomized controlled trial (RCT) with respect to a given intervention, comparator, and outcome of interest, e.g., inferring if an article provides evidence supporting the use of aspirin to reduce risk of stroke, as compared to placebo. We present a new corpus for this task comprising 10,000+ prompts coupled with full-text articles describing RCTs. Results using a suite of models --- ranging from heuristic (rule-based) approaches to attentive neural architectures --- demonstrate the difficulty of the task, which we believe largely owes to the lengthy, technical input texts. To facilitate further work on this important, challenging problem we make the corpus, documentation, a website and leaderboard, and code for baselines and evaluation available at http://evidence-inference.ebm-nlp.com/.

cs.CL↗

arXiv2019

Structured Neural Topic Models for Reviews

We present Variational Aspect-based Latent Topic Allocation (VALTA), a family of autoencoding topic models that learn aspect-based representations of reviews. VALTA defines a user-item encoder that maps bag-of-words vectors for combined reviews associated with each paired user and item onto structured embeddings, which in turn define per-aspect topic weights. We model individual reviews in a structured manner by inferring an aspect assignment for each sentence in a given review, where the per-aspect topic weights obtained by the user-item encoder serve to define a mixture over topics, conditioned on the aspect. The result is an autoencoding neural topic model for reviews, which can be trained in a fully unsupervised manner to learn topics that are structured into aspects. Experimental evaluation on large number of datasets demonstrates that aspects are interpretable, yield higher coherence scores than non-structured autoencoding topic model variants, and can be utilized to perform aspect-based comparison and genre discovery.

cs.CL↗

arXiv2018

Structured Multi-Label Biomedical Text Tagging via Attentive Neural Tree Decoding

We propose a model for tagging unstructured texts with an arbitrary number of terms drawn from a tree-structured vocabulary (i.e., an ontology). We treat this as a special case of sequence-to-sequence learning in which the decoder begins at the root node of an ontological tree and recursively elects to expand child nodes as a function of the input text, the current node, and the latent decoder state. In our experiments the proposed method outperforms state-of-the-art approaches on the important task of automatically assigning MeSH terms to biomedical abstracts.

cs.IR↗

arXiv2018

Learning Disentangled Representations of Texts with Application to Biomedical Abstracts

We propose a method for learning disentangled representations of texts that code for distinct and complementary aspects, with the aim of affording efficient model transfer and interpretability. To induce disentangled embeddings, we propose an adversarial objective based on the (dis)similarity between triplets of documents with respect to specific aspects. Our motivating application is embedding biomedical abstracts describing clinical trials in a manner that disentangles the populations, interventions, and outcomes in a given trial. We show that our method learns representations that encode these clinically salient aspects, and that these can be effectively used to perform aspect-specific retrieval. We demonstrate that the approach generalizes beyond our motivating application in experiments on two multi-aspect review corpora.

cs.CL↗

arXiv2018

A Corpus with Multi-Level Annotations of Patients, Interventions and Outcomes to Support Language Processing for Medical Literature

We present a corpus of 5,000 richly annotated abstracts of medical articles describing clinical randomized controlled trials. Annotations include demarcations of text spans that describe the Patient population enrolled, the Interventions studied and to what they were Compared, and the Outcomes measured (the `PICO' elements). These spans are further annotated at a more granular level, e.g., individual interventions within them are marked and mapped onto a structured medical vocabulary. We acquired annotations from a diverse set of workers with varying levels of expertise and cost. We describe our data collection process and the corpus itself in detail. We then outline a set of challenging NLP tasks that would aid searching of the medical literature and the practice of evidence-based medicine.

cs.CL↗

arXiv2017

Retrofitting Concept Vector Representations of Medical Concepts to Improve Estimates of Semantic Similarity and Relatedness

Estimation of semantic similarity and relatedness between biomedical concepts has utility for many informatics applications. Automated methods fall into two categories: methods based on distributional statistics drawn from text corpora, and methods using the structure of existing knowledge resources. Methods in the former category disregard taxonomic structure, while those in the latter fail to consider semantically relevant empirical information. In this paper, we present a method that retrofits distributional context vector representations of biomedical concepts using structural information from the UMLS Metathesaurus, such that the similarity between vector representations of linked concepts is augmented. We evaluated it on the UMNSRS benchmark. Our results demonstrate that retrofitting of concept vector representations leads to better correlation with human raters for both similarity and relatedness, surpassing the best results reported to date. They also demonstrate a clear improvement in performance on this reference standard for retrofitted vector representations, as compared to those without retrofitting.

cs.CL↗

arXiv2017

Quantifying Mental Health from Social Media with Neural User Embeddings

Mental illnesses adversely affect a significant proportion of the population worldwide. However, the methods traditionally used for estimating and characterizing the prevalence of mental health conditions are time-consuming and expensive. Consequently, best-available estimates concerning the prevalence of mental health conditions are often years out of date. Automated approaches to supplement these survey methods with broad, aggregated information derived from social media content provides a potential means for near real-time estimates at scale. These may, in turn, provide grist for supporting, evaluating and iteratively improving upon public health programs and interventions. We propose a novel model for automated mental health status quantification that incorporates user embeddings. This builds upon recent work exploring representation learning methods that induce embeddings by leveraging social media post histories. Such embeddings capture latent characteristics of individuals (e.g., political leanings) and encode a soft notion of homophily. In this paper, we investigate whether user embeddings learned from twitter post histories encode information that correlates with mental health statuses. To this end, we estimated user embeddings for a set of users known to be affected by depression and post-traumatic stress disorder (PTSD), and for a set of demographically matched `control' users. We then evaluated these embeddings with respect to: (i) their ability to capture homophilic relations with respect to mental health status; and (ii) the performance of downstream mental health prediction models based on these features. Our experimental results demonstrate that the user embeddings capture similarities between users with respect to mental conditions, and are predictive of mental health.

cs.CL↗

arXiv2017

Exploiting Domain Knowledge via Grouped Weight Sharing with Application to Text Categorization

A fundamental advantage of neural models for NLP is their ability to learn representations from scratch. However, in practice this often means ignoring existing external linguistic resources, e.g., WordNet or domain specific ontologies such as the Unified Medical Language System (UMLS). We propose a general, novel method for exploiting such resources via weight sharing. Prior work on weight sharing in neural networks has considered it largely as a means of model compression. In contrast, we treat weight sharing as a flexible mechanism for incorporating prior knowledge into neural models. We show that this approach consistently yields improved performance on classification tasks compared to baseline strategies that do not exploit weight sharing.

cs.CL↗

arXiv2017

Neural Information Retrieval: A Literature Review

A recent "third wave" of Neural Network (NN) approaches now delivers state-of-the-art performance in many machine learning tasks, spanning speech recognition, computer vision, and natural language processing. Because these modern NNs often comprise multiple interconnected layers, this new NN research is often referred to as deep learning. Stemming from this tide of NN work, a number of researchers have recently begun to investigate NN approaches to Information Retrieval (IR). While deep NNs have yet to achieve the same level of success in IR as seen in other areas, the recent surge of interest and work in NNs for IR suggest that this state of affairs may be quickly changing. In this work, we survey the current landscape of Neural IR research, paying special attention to the use of learned representations of queries and documents (i.e., neural embeddings). We highlight the successes of neural IR thus far, catalog obstacles to its wider adoption, and suggest potentially promising directions for future research.

cs.IR↗

arXiv2016

Active Discriminative Text Representation Learning

We propose a new active learning (AL) method for text classification with convolutional neural networks (CNNs). In AL, one selects the instances to be manually labeled with the aim of maximizing model performance with minimal effort. Neural models capitalize on word embeddings as representations (features), tuning these to the task at hand. We argue that AL strategies for multi-layered neural models should focus on selecting instances that most affect the embedding space (i.e., induce discriminative word representations). This is in contrast to traditional AL approaches (e.g., entropy-based uncertainty sampling), which specify higher level objectives. We propose a simple approach for sentence classification that selects instances containing words whose embeddings are likely to be updated with the greatest magnitude, thereby rapidly learning discriminative, task-specific embeddings. We extend this approach to document classification by jointly considering: (1) the expected changes to the constituent word representations; and (2) the model's current overall uncertainty regarding the instance. The relative emphasis placed on these criteria is governed by a stochastic process that favors selecting instances likely to improve representations at the outset of learning, and then shifts toward general uncertainty sampling as AL progresses. Empirical results show that our method outperforms baseline AL approaches on both sentence and document classification tasks. We also show that, as expected, the method quickly learns discriminative word embeddings. To the best of our knowledge, this is the first work on AL addressing neural models for text classification.

cs.CL↗

arXiv2016

Rationale-Augmented Convolutional Neural Networks for Text Classification

We present a new Convolutional Neural Network (CNN) model for text classification that jointly exploits labels on documents and their component sentences. Specifically, we consider scenarios in which annotators explicitly mark sentences (or snippets) that support their overall document categorization, i.e., they provide rationales. Our model exploits such supervision via a hierarchical approach in which each document is represented by a linear combination of the vector representations of its component sentences. We propose a sentence-level convolutional model that estimates the probability that a given sentence is a rationale, and we then scale the contribution of each sentence to the aggregate document representation in proportion to these estimates. Experiments on five classification datasets that have document labels and associated rationales demonstrate that our approach consistently outperforms strong baselines. Moreover, our model naturally provides explanations for its predictions.

cs.CL↗

arXiv2016

Crowdsourcing Information Extraction for Biomedical Systematic Reviews

Information extraction is a critical step in the practice of conducting biomedical systematic literature reviews. Extracted structured data can be aggregated via methods such as statistical meta-analysis. Typically highly trained domain experts extract data for systematic reviews. The high expense of conducting biomedical systematic reviews has motivated researchers to explore lower cost methods that achieve similar rigor without compromising quality. Crowdsourcing represents one such promising approach. In this work-in-progress study, we designed a crowdsourcing task for biomedical information extraction. We briefly report the iterative design process and the results of two pilot testings. We found that giving more concrete examples in the task instruction can help workers better understand the task, especially for concepts that are abstract and confusing. We found a few workers completed most of the work, and our payment level appeared more attractive to workers from low-income countries. In the future, we will further evaluate our results with reference to gold standard extractions, thus assessing the feasibility of tasking crowd workers with extracting biomedical intervention information for systematic reviews.

cs.HC↗