arXiv subjects

Andrew Moore

Publications and source records attributed to Andrew Moore.

17 recordsLinked to original sources

arXiv2026

AT 2020afjz (TSS2020a): The First Fast Extragalactic Transient Discovered by TESS

We report the discovery of AT 2020afjz (TSS2020a): the first hour-scale extragalactic transient discovered in optical wavelengths whose complete evolution -- from explosion onset to decay -- is temporally resolved, and the first such transient discovered by TESS. AT 2020afjz was identified as a $>10\sigma$ detection in the pilot HiLaTS program run within the TESSELLATE Sky Survey, which blindly searches for transient phenomena in TESS data with the TESSELLATE pipeline. Through cross-matching with legacy imaging, we associate it with DES J042144.37$-$383311.3, a member of an interacting galaxy pair at $z_{\rm phot}=0.67^{+0.07}_{-0.10}$. While AT 2020afjz is similar in duration and brightness to GRB afterglows, it exhibits a slow rise time of $\sim1$ hr and lasts for only 2.4 hr above the half-max brightness; modeling the TESS light curve with VegasAfterglow finds that it is best described as either an on-axis "dirty-fireball" or off-axis orphan afterglow. Each of these rare classifications hinge upon a non-detection at gamma-ray energies, but as Fermi-GBM was Earth-occulted at the time of explosion, AT 2020afjz's gamma-quiet nature cannot be definitively confirmed. Regardless, AT 2020afjz demonstrates TESS's power to discover fast extragalactic transients, and heralds a new population awaiting discovery with TESSELLATE.

astro-ph.HE↗

arXiv2026

Creating a Hybrid Rule and Neural Network Based Semantic Tagger using Silver Standard Data: the PyMUSAS framework for Multilingual Semantic Annotation

Word Sense Disambiguation (WSD) has been widely evaluated using the semantic frameworks of WordNet, BabelNet, and the Oxford Dictionary of English. However, for the UCREL Semantic Analysis System (USAS) framework, no open extensive evaluation has been performed beyond lexical coverage or single language evaluation. In this work, we perform the largest semantic tagging evaluation of the rule based system that uses the lexical resources in the USAS framework covering five different languages using four existing datasets and one novel Chinese dataset. We create a new silver labelled English dataset, to overcome the lack of manually tagged training data, that we train and evaluate various mono and multilingual neural models in both mono and cross-lingual evaluation setups with comparisons to their rule based counterparts, and show how a rule based system can be enhanced with a neural network model. The resulting neural network models, including the data they were trained on, the Chinese evaluation dataset, and all of the code have been released as open resources.

cs.CL↗

arXiv2025

HealthcareNLP: where are we and what is next?

This proposed tutorial focuses on Healthcare Domain Applications of NLP, what we have achieved around HealthcareNLP, and the challenges that lie ahead for the future. Existing reviews in this domain either overlook some important tasks, such as synthetic data generation for addressing privacy concerns, or explainable clinical NLP for improved integration and implementation, or fail to mention important methodologies, including retrieval augmented generation and the neural symbolic integration of LLMs and KGs. In light of this, the goal of this tutorial is to provide an introductory overview of the most important sub-areas of a patient- and resource-oriented HealthcareNLP, with three layers of hierarchy: data/resource layer: annotation guidelines, ethical approvals, governance, synthetic data; NLP-Eval layer: NLP tasks such as NER, RE, sentiment analysis, and linking/coding with categorised methods, leading to explainable HealthAI; patients layer: Patient Public Involvement and Engagement (PPIE), health literacy, translation, simplification, and summarisation (also NLP tasks), and shared decision-making support. A hands-on session will be included in the tutorial for the audience to use HealthcareNLP applications. The target audience includes NLP practitioners in the healthcare application domain, NLP researchers who are interested in domain applications, healthcare researchers, and students from NLP fields. The type of tutorial is "Introductory to CL/NLP topics (HealthcareNLP)" and the audience does not need prior knowledge to attend this. Tutorial materials: https://github.com/4dpicture/HealthNLP

cs.CL↗

arXiv2025

TESSELLATE: Piecing Together the Variable Sky With TESS

We present TESSELLATE, a dedicated pipeline for performing an untargeted search documenting all variable phenomena captured by the TESS space telescope. Building on the TESSreduce difference imaging pipeline, TESSELLATE extracts calibrated and reduced photometric data for every full frame image in the TESS archive. Using this data, we systematically identify transient, variable and non-sidereal signals across timescales ranging from minutes to weeks. The high cadence and wide field of view of TESS enables us to conduct a comprehensive search of the entire sky to a depth of ~17 $m_i$. Based on the volumetric rates for known fast transients, we expect there to be numerous Fast Blue Optical Transients and Gamma Ray Burst afterglows present in the existing TESS dataset. Beyond transients, TESSELLATE can also identify new variable stars and exoplanet candidates, and recover known asteroids. We classify events using machine learning techniques and the work of citizen scientists via the Zooniverse Cosmic Cataclysms project. Finally, we introduce the TESSELLATE Sky Survey: a complete, open catalog of the variable sky observed by TESS.

astro-ph.IM↗

arXiv2020

Multi-task Learning of Negation and Speculation for Targeted Sentiment Classification

The majority of work in targeted sentiment analysis has concentrated on finding better methods to improve the overall results. Within this paper we show that these models are not robust to linguistic phenomena, specifically negation and speculation. In this paper, we propose a multi-task learning method to incorporate information from syntactic and semantic auxiliary tasks, including negation and speculation scope detection, to create English-language models that are more robust to these phenomena. Further we create two challenge datasets to evaluate model performance on negated and speculative samples. We find that multi-task models and transfer learning via language modelling can improve performance on these challenge datasets, but the overall performances indicate that there is still much room for improvement. We release both the datasets and the source code at https://github.com/jerbarnes/multitask_negation_for_targeted_sentiment.

cs.CL↗

arXiv2019

FIESTA: Fast IdEntification of State-of-The-Art models using adaptive bandit algorithms

We present FIESTA, a model selection approach that significantly reduces the computational resources required to reliably identify state-of-the-art performance from large collections of candidate models. Despite being known to produce unreliable comparisons, it is still common practice to compare model evaluations based on single choices of random seeds. We show that reliable model selection also requires evaluations based on multiple train-test splits (contrary to common practice in many shared tasks). Using bandit theory from the statistics literature, we are able to adaptively determine appropriate numbers of data splits and random seeds used to evaluate each model, focusing computational resources on the evaluation of promising models whilst avoiding wasting evaluations on models with lower performance. Furthermore, our user-friendly Python implementation produces confidence guarantees of correctly selecting the optimal model. We evaluate our algorithms by selecting between 8 target-dependent sentiment analysis methods using dramatically fewer model evaluations than current model selection approaches.

cs.LG↗

arXiv2018

Bringing replication and reproduction together with generalisability in NLP: Three reproduction studies for Target Dependent Sentiment Analysis

Lack of repeatability and generalisability are two significant threats to continuing scientific development in Natural Language Processing. Language models and learning methods are so complex that scientific conference papers no longer contain enough space for the technical depth required for replication or reproduction. Taking Target Dependent Sentiment Analysis as a case study, we show how recent work in the field has not consistently released code, or described settings for learning methods in enough detail, and lacks comparability and generalisability in train, test or validation data. To investigate generalisability and to enable state of the art comparative evaluations, we carry out the first reproduction studies of three groups of complementary methods and perform the first large-scale mass evaluation on six different English datasets. Reflecting on our experiences, we recommend that future replication or reproduction experiments should always consider a variety of datasets alongside documenting and releasing their methods and published code in order to minimise the barriers to both repeatability and generalisability. We have released our code with a model zoo on GitHub with Jupyter Notebooks to aid understanding and full documentation, and we recommend that others do the same with their papers at submission time through an anonymised GitHub account.

cs.CL↗

arXiv2017

Lancaster A at SemEval-2017 Task 5: Evaluation metrics matter: predicting sentiment from financial news headlines

This paper describes our participation in Task 5 track 2 of SemEval 2017 to predict the sentiment of financial news headlines for a specific company on a continuous scale between -1 and 1. We tackled the problem using a number of approaches, utilising a Support Vector Regression (SVR) and a Bidirectional Long Short-Term Memory (BLSTM). We found an improvement of 4-6% using the LSTM model over the SVR and came fourth in the track. We report a number of different evaluations using a finance specific word embedding model and reflect on the effects of using different evaluation metrics.

cs.CL↗

arXiv2013

Mix-nets: Factored Mixtures of Gaussians in Bayesian Networks With Mixed Continuous And Discrete Variables

Recently developed techniques have made it possible to quickly learn accurate probability density functions from data in low-dimensional continuous space. In particular, mixtures of Gaussians can be fitted to data very quickly using an accelerated EM algorithm that employs multiresolution kd-trees (Moore, 1999). In this paper, we propose a kind of Bayesian networks in which low-dimensional mixtures of Gaussians over different subsets of the domain's variables are combined into a coherent joint probability model over the entire domain. The network is also capable of modeling complex dependencies between discrete variables and continuous variables without requiring discretization of the continuous variables. We present efficient heuristic algorithms for automatically learning these networks from data, and perform comparative experiments illustrated how well these networks model real scientific data and synthetic data. We also briefly discuss some possible improvements to the networks, as well as possible applications.

cs.LG↗

arXiv2013

The Anchors Hierachy: Using the triangle inequality to survive high dimensional data

This paper is about metric data structures in high-dimensional or non-Euclidean space that permit cached sufficient statistics accelerations of learning algorithms. It has recently been shown that for less than about 10 dimensions, decorating kd-trees with additional "cached sufficient statistics" such as first and second moments and contingency tables can provide satisfying acceleration for a very wide range of statistical learning tasks such as kernel regression, locally weighted regression, k-means clustering, mixture modeling and Bayes Net learning. In this paper, we begin by defining the anchors hierarchy - a fast data structure and algorithm for localizing data based only on a triangle-inequality-obeying distance metric. We show how this, in its own right, gives a fast and effective clustering of data. But more importantly we show how it can produce a well-balanced structure similar to a Ball-Tree (Omohundro, 1991) or a kind of metric tree (Uhlmann, 1991; Ciaccia, Patella, & Zezula, 1997) in a way that is neither "top-down" nor "bottom-up" but instead "middle-out". We then show how this structure, decorated with cached sufficient statistics, allows a wide variety of statistical learning algorithms to be accelerated even in thousands of dimensions.

cs.LG↗

arXiv2012

Interpolating Conditional Density Trees

Joint distributions over many variables are frequently modeled by decomposing them into products of simpler, lower-dimensional conditional distributions, such as in sparsely connected Bayesian networks. However, automatically learning such models can be very computationally expensive when there are many datapoints and many continuous variables with complex nonlinear relationships, particularly when no good ways of decomposing the joint distribution are known a priori. In such situations, previous research has generally focused on the use of discretization techniques in which each continuous variable has a single discretization that is used throughout the entire network. \ In this paper, we present and compare a wide variety of tree-based algorithms for learning and evaluating conditional density estimates over continuous variables. These trees can be thought of as discretizations that vary according to the particular interactions being modeled; however, the density within a given leaf of the tree need not be assumed constant, and we show that such nonuniform leaf densities lead to more accurate density estimation. We have developed Bayesian network structure-learning algorithms that employ these tree-based conditional density representations, and we show that they can be used to practically learn complex joint probability models over dozens of continuous variables from thousands of datapoints. We focus on finding models that are simultaneously accurate, fast to learn, and fast to evaluate once they are learned.

cs.LG↗

arXiv2012

Real-valued All-Dimensions search: Low-overhead rapid searching over subsets of attributes

This paper is about searching the combinatorial space of contingency tables during the inner loop of a nonlinear statistical optimization. Examples of this operation in various data analytic communities include searching for nonlinear combinations of attributes that contribute significantly to a regression (Statistics), searching for items to include in a decision list (machine learning) and association rule hunting (Data Mining). This paper investigates a new, efficient approach to this class of problems, called RADSEARCH (Real-valued All-Dimensions-tree Search). RADSEARCH finds the global optimum, and this gives us the opportunity to empirically evaluate the question: apart from algorithmic elegance what does this attention to optimality buy us? We compare RADSEARCH with other recent successful search algorithms such as CN2, PRIM, APriori, OPUS and DenseMiner. Finally, we introduce RADREG, a new regression algorithm for learning real-valued outputs based on RADSEARCHing for high-order interactions.

cs.AI↗

arXiv2012

A Tractable Approach to Finding Closest Truncated-commute-time Neighbors in Large Graphs

Recently there has been much interest in graph-based learning, with applications in collaborative filtering for recommender networks, link prediction for social networks and fraud detection. These networks can consist of millions of entities, and so it is very important to develop highly efficient techniques. We are especially interested in accelerating random walk approaches to compute some very interesting proximity measures of these kinds of graphs. These measures have been shown to do well empirically (Liben-Nowell & Kleinberg, 2003; Brand, 2005). We introduce a truncated variation on a well-known measure, namely commute times arising from random walks on graphs. We present a very novel algorithm to compute all interesting pairs of approximate nearest neighbors in truncated commute times, without computing it between all pairs. We show results on both simulated and real graphs of size up to 100; 000 entities, which indicate near-linear scaling in computation time.

cs.SI↗

arXiv2007

Efficient intra- and inter-night linking of asteroid detections using kd-trees

The Panoramic Survey Telescope And Rapid Response System (Pan-STARRS) under development at the University of Hawaii's Institute for Astronomy is creating the first fully automated end-to-end Moving Object Processing System (MOPS) in the world. It will be capable of identifying detections of moving objects in our solar system and linking those detections within and between nights, attributing those detections to known objects, calculating initial and differentially-corrected orbits for linked detections, precovering detections when they exist, and orbit identification. Here we describe new kd-tree and variable-tree algorithms that allow fast, efficient, scalable linking of intra and inter-night detections. Using a pseudo-realistic simulation of the Pan-STARRS survey strategy incorporating weather, astrometric accuracy and false detections we have achieved nearly 100% efficiency and accuracy for intra-night linking and nearly 100% efficiency for inter-night linking within a lunation. At realistic sky-plane densities for both real and false detections the intra-night linking of detections into `tracks' currently has an accuracy of 0.3%. Successful tests of the MOPS on real source detections from the Spacewatch asteroid survey indicate that the MOPS is capable of identifying asteroids in real data.

astro-ph↗

arXiv2005

Massive Science with VO and Grids

There is a growing need for massive computational resources for the analysis of new astronomical datasets. To tackle this problem, we present here our first steps towards marrying two new and emerging technologies; the Virtual Observatory (e.g, AstroGrid) and the computational grid (e.g. TeraGrid, COSMOS etc.). We discuss the construction of VOTechBroker, which is a modular software tool designed to abstract the tasks of submission and management of a large number of computational jobs to a distributed computer system. The broker will also interact with the AstroGrid workflow and MySpace environments. We discuss our planned usages of the VOTechBroker in computing a huge number of n-point correlation functions from the SDSS data and massive model-fitting of millions of CMBfast models to WMAP data. We also discuss other applications including the determination of the XMM Cluster Survey selection function and the construction of new WMAP maps.

astro-ph↗

arXiv2001

Controlling the False Discovery Rate in Astrophysical Data Analysis

The False Discovery Rate (FDR) is a new statistical procedure to control the number of mistakes made when performing multiple hypothesis tests, i.e. when comparing many data against a given model hypothesis. The key advantage of FDR is that it allows one to a priori control the average fraction of false rejections made (when comparing to the null hypothesis) over the total number of rejections performed. We compare FDR to the standard procedure of rejecting all tests that do not match the null hypothesis above some arbitrarily chosen confidence limit, e.g. 2 sigma, or at the 95% confidence level. When using FDR, we find a similar rate of correct detections, but with significantly fewer false detections. Moreover, the FDR procedure is quick and easy to compute and can be trivially adapted to work with correlated data. The purpose of this paper is to introduce the FDR procedure to the astrophysics community. We illustrate the power of FDR through several astronomical examples, including the detection of features against a smooth one-dimensional function, e.g. seeing the ``baryon wiggles'' in a power spectrum of matter fluctuations, and source pixel detection in imaging data. In this era of large datasets and high precision measurements, FDR provides the means to adaptively control a scientifically meaningful quantity -- the number of false discoveries made when conducting multiple hypothesis tests.

astro-ph↗

arXiv2000

Fast Algorithms and Efficient Statistics: N-point Correlation Functions

We present here a new algorithm for the fast computation of N-point correlation functions in large astronomical data sets. The algorithm is based on kdtrees which are decorated with cached sufficient statistics thus allowing for orders of magnitude speed-ups over the naive non-tree-based implementation of correlation functions. We further discuss the use of controlled approximations within the computation which allows for further acceleration. In summary, our algorithm now makes it possible to compute exact, all-pairs, measurements of the 2, 3 and 4-point correlation functions for cosmological data sets like the Sloan Digital Sky Survey (SDSS; York et al. 2000) and the next generation of Cosmic Microwave Background experiments (see Szapudi et al. 2000).

astro-ph↗