arXiv subjects

Cuong Nguyen

Publications and source records attributed to Cuong Nguyen.

30 records · Page 2Linked to original sources

arXiv2023

Task Weighting in Meta-learning with Trajectory Optimisation

Developing meta-learning algorithms that are un-biased toward a subset of training tasks often requires hand-designed criteria to weight tasks, potentially resulting in sub-optimal solutions. In this paper, we introduce a new principled and fully-automated task-weighting algorithm for meta-learning methods. By considering the weights of tasks within the same mini-batch as an action, and the meta-parameter of interest as the system state, we cast the task-weighting meta-learning problem to a trajectory optimisation and employ the iterative linear quadratic regulator to determine the optimal action or weights of tasks. We theoretically show that the proposed algorithm converges to an $ε_{0}$-stationary point, and empirically demonstrate that the proposed approach out-performs common hand-engineering weighting methods in two few-shot learning benchmarks.

cs.LG↗

arXiv2023

Geo-mechanical aspects for breakage detachment of rock fines by Darcys flow

Suspension-colloidal-nano transport in porous media encompasses the detachment of detrital fines against electrostatic attraction and authigenic fines by breakage, from the rock surface. While much is currently known about the underlying mechanisms governing detachment of detrital particles, including detachment criteria at the pore scale and its upscaling for the core scale, a critical gap exists due to absence of this knowledge for authigenic fines. Integrating 3D Timoshenkos beam theory of elastic cylinder deformation with CFD-based model for viscous flow around the attached particle and with strength failure criteria for particle-rock bond, we developed a novel theory for fines detachment by breakage at the pore scale. The breakage criterium derived includes analytical expressions for tensile and shear stress maxima along with two geometric diagrams which allow determining the breaking stress. This leads to an explicit formula for the breakage flow velocity. Its upscaling yields a mathematical model for fines detachment by breakage, expressed in the form of the maximum retained concentration of attached fines versus flow velocity -- maximum retention function (MRF) for breakage. We performed corefloods with piecewise constant increasing flow rates, measuring breakthrough concentration and pressure drop across the core. The behaviour of the measured data is consistent with two-population colloidal transport, attributed to detrital and authigenic fines migration. Indeed, the laboratory data show high match with the analytical model for two-population colloidal transport, which validates the proposed mathematical model for fines detachment by breakage.

physics.geo-ph↗

arXiv2022

Instance-Dependent Noisy Label Learning via Graphical Modelling

Noisy labels are unavoidable yet troublesome in the ecosystem of deep learning because models can easily overfit them. There are many types of label noise, such as symmetric, asymmetric and instance-dependent noise (IDN), with IDN being the only type that depends on image information. Such dependence on image information makes IDN a critical type of label noise to study, given that labelling mistakes are caused in large part by insufficient or ambiguous information about the visual classes present in images. Aiming to provide an effective technique to address IDN, we present a new graphical modelling approach called InstanceGM, that combines discriminative and generative models. The main contributions of InstanceGM are: i) the use of the continuous Bernoulli distribution to train the generative model, offering significant training advantages, and ii) the exploration of a state-of-the-art noisy-label discriminative classifier to generate clean labels from instance-dependent noisy-label samples. InstanceGM is competitive with current noisy-label learning approaches, particularly in IDN benchmarks using synthetic and real-world datasets, where our method shows better accuracy than the competitors in most experiments.

cs.CV↗

arXiv2022

Why So Inflammatory? Explainability in Automatic Detection of Inflammatory Social Media Users

Hate speech and misinformation, spread over social networking services (SNS) such as Facebook and Twitter, have inflamed ethnic and political violence in countries across the globe. We argue that there is limited research on this problem within the context of the Global South and present an approach for tackling them. Prior works have shown how machine learning models built with user-level interaction features can effectively identify users who spread inflammatory content. While this technique is beneficial in low-resource language settings where linguistic resources such as ground truth data and processing capabilities are lacking, it is still unclear how these interaction features contribute to model performance. In this work, we investigate and show significant differences in interaction features between users who spread inflammatory content and others who do not, applying explainability tools to understand our trained model. We find that features with higher interaction significance (such as account age and activity count) show higher explanatory power than features with lower interaction significance (such as name length and if the user has a location on their bio). Our work extends research directions that aim to understand the nature of inflammatory content in low-resource, high-risk contexts as the growth of social media use in the Global South outstrips moderation efforts.

cs.SI↗

arXiv2021

PAC-Bayes meta-learning with implicit task-specific posteriors

We introduce a new and rigorously-formulated PAC-Bayes meta-learning algorithm that solves few-shot learning. Our proposed method extends the PAC-Bayes framework from a single task setting to the meta-learning multiple task setting to upper-bound the error evaluated on any, even unseen, tasks and samples. We also propose a generative-based approach to estimate the posterior of task-specific model parameters more expressively compared to the usual assumption based on a multivariate normal distribution with a diagonal covariance matrix. We show that the models trained with our proposed meta-learning algorithm are well calibrated and accurate, with state-of-the-art calibration and classification results on few-shot classification (mini-ImageNet and tiered-ImageNet) and regression (multi-modal task-distribution regression) benchmarks.

cs.LG↗

arXiv2021

Semi-supervised classification of radiology images with NoTeacher: A Teacher that is not Mean

Deep learning models achieve strong performance for radiology image classification, but their practical application is bottlenecked by the need for large labeled training datasets. Semi-supervised learning (SSL) approaches leverage small labeled datasets alongside larger unlabeled datasets and offer potential for reducing labeling cost. In this work, we introduce NoTeacher, a novel consistency-based SSL framework which incorporates probabilistic graphical models. Unlike Mean Teacher which maintains a teacher network updated via a temporal ensemble, NoTeacher employs two independent networks, thereby eliminating the need for a teacher network. We demonstrate how NoTeacher can be customized to handle a range of challenges in radiology image classification. Specifically, we describe adaptations for scenarios with 2D and 3D inputs, uni and multi-label classification, and class distribution mismatch between labeled and unlabeled portions of the training data. In realistic empirical evaluations on three public benchmark datasets spanning the workhorse modalities of radiology (X-Ray, CT, MRI), we show that NoTeacher achieves over 90-95% of the fully supervised AUROC with less than 5-15% labeling budget. Further, NoTeacher outperforms established SSL methods with minimal hyperparameter tuning, and has implications as a principled and practical option for semisupervised learning in radiology applications.

cs.CV↗

arXiv2021

Approximate Bayesian Computation for an Explicit-Duration Hidden Markov Model of COVID-19 Hospital Trajectories

We address the problem of modeling constrained hospital resources in the midst of the COVID-19 pandemic in order to inform decision-makers of future demand and assess the societal value of possible interventions. For broad applicability, we focus on the common yet challenging scenario where patient-level data for a region of interest are not available. Instead, given daily admissions counts, we model aggregated counts of observed resource use, such as the number of patients in the general ward, in the intensive care unit, or on a ventilator. In order to explain how individual patient trajectories produce these counts, we propose an aggregate count explicit-duration hidden Markov model, nicknamed the ACED-HMM, with an interpretable, compact parameterization. We develop an Approximate Bayesian Computation approach that draws samples from the posterior distribution over the model's transition and duration parameters given aggregate counts from a specific location, thus adapting the model to a region or individual hospital site of interest. Samples from this posterior can then be used to produce future forecasts of any counts of interest. Using data from the United States and the United Kingdom, we show our mechanistic approach provides competitive probabilistic forecasts for the future even as the dynamics of the pandemic shift. Furthermore, we show how our model provides insight about recovery probabilities or length of stay distributions, and we suggest its potential to answer challenging what-if questions about the societal value of possible interventions.

stat.AP↗

arXiv2021

Similarity of Classification Tasks

Recent advances in meta-learning has led to remarkable performances on several few-shot learning benchmarks. However, such success often ignores the similarity between training and testing tasks, resulting in a potential bias evaluation. We, therefore, propose a generative approach based on a variant of Latent Dirichlet Allocation to analyse task similarity to optimise and better understand the performance of meta-learning. We demonstrate that the proposed method can provide an insightful evaluation for meta-learning algorithms on two few-shot classification benchmarks that matches common intuition: the more similar the higher performance. Based on this similarity measure, we propose a task-selection strategy for meta-learning and show that it can produce more accurate classification results than methods that randomly select training tasks.

cs.LG↗

arXiv2019

Uncertainty in Model-Agnostic Meta-Learning using Variational Inference

We introduce a new, rigorously-formulated Bayesian meta-learning algorithm that learns a probability distribution of model parameter prior for few-shot learning. The proposed algorithm employs a gradient-based variational inference to infer the posterior of model parameters to a new task. Our algorithm can be applied to any model architecture and can be implemented in various machine learning paradigms, including regression and classification. We show that the models trained with our proposed meta-learning algorithm are well calibrated and accurate, with state-of-the-art calibration and classification results on two few-shot classification benchmarks (Omniglot and Mini-ImageNet), and competitive results in a multi-modal task-distribution regression.

cs.LG↗

arXiv2019

Unsupervised Task Design to Meta-Train Medical Image Classifiers

Meta-training has been empirically demonstrated to be the most effective pre-training method for few-shot learning of medical image classifiers (i.e., classifiers modeled with small training sets). However, the effectiveness of meta-training relies on the availability of a reasonable number of hand-designed classification tasks, which are costly to obtain, and consequently rarely available. In this paper, we propose a new method to unsupervisedly design a large number of classification tasks to meta-train medical image classifiers. We evaluate our method on a breast dynamically contrast enhanced magnetic resonance imaging (DCE-MRI) data set that has been used to benchmark few-shot training methods of medical image classifiers. Our results show that the proposed unsupervised task design to meta-train medical image classifiers builds a pre-trained model that, after fine-tuning, produces better classification results than other unsupervised and supervised pre-training methods, and competitive results with respect to meta-training that relies on hand-designed classification tasks.

cs.CV↗

arXiv2018

Onion Curve: A Space Filling Curve with Near-Optimal Clustering

Space filling curves (SFCs) are widely used in the design of indexes for spatial and temporal data. Clustering is a key metric for an SFC, that measures how well the curve preserves locality in moving from higher dimensions to a single dimension. We present the {\em onion curve}, an SFC whose clustering performance is provably close to optimal for the cube and near-cube shaped query sets, irrespective of the side length of the query. We show that in contrast, the clustering performance of the widely used Hilbert curve can be far from optimal, even for cube-shaped queries. Since the clustering performance of an SFC is critical to the efficiency of multi-dimensional indexes based on the SFC, the onion curve can deliver improved performance for data structures involving multi-dimensional data.

cs.CG↗

arXiv2016

Bridging the Gap between Crosslinking Chemistry and Directed Assembly of Metasurfaces Using Electrohydrodynamic Flow

Advances in understanding chemical and physical driving forces in self-assembly allow the fabrication of unique nanoarchitectures with subwavelength building blocks as the basis for plasmonic and metamaterial devices. Chemical crosslinking of colloidal nanospheres has produced among the smallest gap spacings, necessary to obtain regions of greatly enhanced electric field, hotspots, which are critical to tailor light-matter interactions. However, obtaining uniform electromagnetic response of dense nanoantennas over large area for use in devices remains challenging. In this work, electrohydrodynamic (EHD) flow and chemical crosslinking is combined to form dense, yet discrete, Au nanosphere clusters (oligomers) on a working electrode. EHD provides a long range driving force to bring nanospheres together and anhydride crosslinking yields 0.9 nm gap spacings. Using selective chemistry, nanospheres are simultaneously crosslinked onto a block copolymer template, producing oligomers with a narrower wavelength band width and higher hotspot intensity than monolayer structures produced without a template. We investigate nanoantenna response via full wave simulations, ultraviolet-visible spectroscopy, and surface enhanced Raman scattering (SERS). Nanoantennas exhibit uniform hotspot intensity and gap spacing. Simulations show field enhancements of 600, correlating well with measured average SERS enhancement of 1.4x10^9. Nanoantenna substrates produce a SERS signal with a relative standard deviation of 10% measured over a 1 mm2 area, crucial for nano-optical devices such as optical sensors, among other applications. Understanding long range (EHD flow) and short range (chemical crosslinking) driving forces provides the control for assembling colloidal nanoparticles in architectures for large area plasmonic and metasurface device fabrication.

cond-mat.soft↗