This paper introduces a new method for SAR imagery region discrimination using geodesic distances.
problem Region discrimination in monopolarized SAR imagery.
method Geodesic distance between GI0 models. result Advantages of using geodesic distance over stochastic distances.
The paper proposes a method to focus on discriminative regions for better unsupervised domain adaptation.
problem Unsupervised domain adaptation with limited target domain labels.
method Probabilistic certainty estimate of regions to focus on during classification.
result State-of-the-art results on various datasets compared to recent methods.
Geotagged data can be used to describe regions in the world and discover local themes. However, not all data produced within a region is necessarily specifically descriptive of that area. To surface the content that is characteristic for a region, we present the geographical hierarchy model (GHM), a probabilistic model…
We introduce the "Energy-based Generative Adversarial Network" model (EBGAN) which views the discriminator as an energy function that attributes low energies to the regions near the data manifold and higher energies to other regions. Similar to the probabilistic GANs, a generator is seen as being trained to produce con…
Developed a multiway classification method for sparse data.
problem Classification of multiway arrays with sparsity.
method Extended Distance Weighted Discrimination (DWD) to multiway context, accounting for sparsity.
result Improves classification accuracy in multiway structured data.
DBNet improves natural language image localization and detection.
problem Natural language-based visual entity localization with limited accuracy.
method Discriminative bimodal neural network (DBNet) trained with extensive negative samples.
result Significantly outperforms previous methods on Visual Genome dataset.
Grad-CAM visualizes CNN predictions to explain model decisions.
problem Making CNN models transparent and understandable.
method Using class-specific gradient information to localize important regions.
result Improved understanding of CNN-based models, including image captioning and VQA.
DNLL loss improves deep LDA accuracy and consistency.
problem Pathological solutions in unconstrained Deep LDA.
method Introducing Discriminative Negative Log-Likelihood (DNLL) loss.
result Deep LDA trained with DNLL produces clean latent spaces and better calibrated probabilities.
NCC is inefficient in higher dimensions, NCDA improves performance.
problem Inefficiency of NCC in higher dimensions.
method Combining NCC with LDA to create NCDA.
result NCDA outperforms NCC and competes with LDA and QDA.
Proposes interpretable attention for video action recognition.
problem Efficient video action recognition with attention mechanisms.
method Spatial-temporal attention mechanism with saliency masks and convolutional LSTM.
result Improves video action recognition accuracy and spatial-temporal localization.
A mixture of experts model predicts brain activation from word stimuli.
problem Classical encoding models ignore connections among brain regions.
method Mixture of experts capturing ROI-specific brain activity patterns.
result Model predicts entire brain activation with high spatial accuracy.
Efficient system classifies EEG signals for cognitive tasks using nuclear features.
problem Classification of raw EEG signals for cognitive tasks is challenging.
method Singular value decomposition for computing dominant variances of EEG signals, using them as nuclear features, and a simple classifier.
result Nuclear features from frontal brain region achieved 100% prediction accuracy.
Develops a method to interpret deep learning models by identifying key features.
problem Revealing the decision-making process of blackbox models from raw data to predictions.
method Adversarial attacks to localize discriminative features with statistical guarantees.
result Locally identified features are both biologically plausible and statistically significant.
The paper analyzes deep neural network classification regions and their decision boundaries.
problem Understanding the geometric properties of deep neural network classifiers.
method Empirical investigation of deep neural networks' classification regions and decision boundaries.
result Deep neural networks learn connected classification regions with flat decision boundaries.
Researchers in functional neuroimaging mostly use activation coordinates to formulate their hypotheses. Instead, we propose to use the full statistical images to define regions of interest (ROIs). This paper presents two machine learning approaches, transfer learning and selection transfer, that are compared upon their…
Machine learning identifies boundaries of real solutions in polynomial systems.
problem Locating boundaries in parameter space for real solutions of polynomial systems.
method Supervised machine learning approach using nearest neighbor and deep learning approximations.
result Efficiently approximates the real discriminant locus for multidimensional parameter spaces.
RPSF improves shapelet-based random forest accuracy and speed.
problem Limitations of random shapelet forest, including high training cost and limited information per node.
method Combines pairs of shapelets, omits threshold searching, and uses DMDI for interpretability.
result Improves accuracy and training speed of shapelet-based random forest.
In recent work on both generative and discriminative score to log-likelihood-ratio calibration, it was shown that linear transforms give good accuracy only for a limited range of operating points. Moreover, these methods required tailoring of the calibration training objective functions in order to target the desired r…
Develops a method to segment high-dimensional data into regions with different local intrinsic dimensions.
problem Data often has varying intrinsic dimensions within the same dataset, challenging traditional analysis.
method Discovers regions with different local intrinsic dimensions and segments the data accordingly.
result Many real-world data sets contain regions with widely heterogeneous dimensions, which can be segmented using local intrinsic dimension.
We propose a new inferential framework for constructing confidence regions and testing hypotheses in statistical models specified by a system of high dimensional estimating equations. We construct an influence function by projecting the fitted estimating equations to a sparse direction obtained by solving a large-scale…
Study identifies regions where scoring rules reliably detect forecast errors.
problem Insufficient reliability of scoring rules in evaluating multivariate probabilistic forecasts.
method Systematic finite-sample analysis of proper scoring rules on synthetic and real-world data.
result Identified regions of reliability for scoring rules in time-series forecasting.
Develops counterfactual visual explanations to show how images could change to classify differently.
problem Creating understandable explanations for vision system predictions.
method Selects a distractor image and identifies spatial regions to modify for different classification.
result Users trained with counterfactual explanations perform better in fine-grained bird classification.
Confidential Guardian prevents model abstention from being used to discriminate.
problem Dishonest institutions can exploit machine learning model abstention to unfairly deny services.
method Confidential Guardian uses zero-knowledge proofs to verify model confidence and detect suppression.
result Confidential Guardian effectively prevents the misuse of cautious predictions.
The vast majority of real world classification problems are imbalanced, meaning there are far fewer data from the class of interest (the positive class) than from other classes. We propose two machine learning algorithms to handle highly imbalanced classification problems. The classifiers constructed by both methods ar…
EBM uses high-dimensional imaging biomarkers to improve dementia progression estimation.
problem Current EBMs only use scalar biomarkers, limiting accuracy from cross-sectional data.
method Proposes nDEBM, a novel method using semi-supervised SVM on voxel-wise imaging biomarkers.
result nDEBM outperforms state-of-the-art EBM methods using regional volume biomarkers.
High-dimensional linear classifiers, such as the support vector machine (SVM) and distance weighted discrimination (DWD), are commonly used in biomedical research to distinguish groups of subjects based on a large number of features. However, their use is limited to applications where a single vector of features is mea…
PromID predicts human promoter positions with high accuracy.
problem Difficulties in identifying human promoter sequences.
method Deep learning approach predicting exact TSS positions.
result Significantly reduces false positive predictions.
Enhances interpolation paths in latent space using particle filters.
problem Generating meaningful interpolations between data points in latent space.
method Introduces a discriminator network to guide particle filter sampling of interpolation paths.
result Improved variability and stronger drift towards high data density areas.
A hierarchy of GNNs based on learnable local features is proposed.
problem Limited understanding of GNN architectures and their systematic construction.
method A hierarchy of GNNs based on aggregation regions is derived, and theoretical results are provided.
result Simple GNN architecture exceeds Weisfeiler-Lehman graph isomorphism test.
Graph embedding improves fMRI classification and reveals brain region differences in ASD.
problem Difficult to embed informative brain fMRI representations due to high dimensionality and low SNR.
method Modelled fMRI as a graph, used GNN to learn from graph data, incorporated mutual information loss (Infomax).
result Infomax graph embedding improves classification performance and reveals separable nodal representations of ASD and HC groups.
Method uses aggregate crop statistics to improve satellite-based crop type mapping.
problem Limited field-level crop labels for training satellite-based maps.
method Corrects classifier by accounting for shifts in crop type composition and feature means.
result Substantial improvements in overall classification accuracy, reducing misclassifications by 21.9% on average.
Spatially-aware metrics improve uncertainty evaluation in segmentation.
problem Uncertainty evaluation metrics treat voxels independently, ignoring spatial context.
method Proposed three spatially aware metrics incorporating structural and boundary information.
result Improved alignment with clinically important factors and better discrimination between uncertainty patterns.
A new deep metric learning method pulls embeddings towards dense clusters to improve classification accuracy.
problem Improving classification accuracy in deep metric learning models.
method Density Aware Metric Learning (DAML) which pulls embeddings towards the densest regions of clusters for each class.
result DAML achieves faster convergence and higher generalizability compared to existing methods.
Paper tackles unsupervised learning under latent label shift across domains.
problem Discovering classes from unlabeled data with shifting label distributions.
method Introduces unsupervised learning under Latent Label Shift (LLS), leveraging domain-discriminative models.
result Proves that with domain information, unsupervised classification can improve upon standard methods.
SaliencyMix augments images with salient patches to improve model generalization.
problem Improving deep learning model generalization through better data augmentation.
method Carefully selects salient patches from images and mixes them with the target image.
result Achieves state-of-the-art top-1 error rates and robustness against adversarial attacks.
BEAMs use adversarial training to improve RBM performance.
problem Likelihood-based training fails to penalize high-probability regions.
method Adversarial training against RBM hidden layer activations.
result BEAMs outperform RBMs and GANs on benchmarks.
Study examines how familiarity of images affects EEG patterns.
problem Understanding brain responses to images of varying familiarity.
method Recorded EEG of 21 subjects viewing images of varying familiarity levels. Features extracted from EEG data, including spatial-statistical, wavelet, frequency, and harmonic parameters. SVM used for classification.
result Best classification rates in pre-frontal and frontal regions, with wavelet, frequency, and harmonic features being most discriminative.
New framework quantifies uncertainty in data and models using RKHS.
problem Quantifying uncertainty in data and models.
method Projecting data into RKHS, transforming PDF, decomposing gradient flow.
result Decomposes uncertainty moments, providing discriminative resolution.
MDGCN improves hyperspectral image classification by dynamically updating graphs.
problem Traditional CNNs struggle with irregular image regions and class boundaries.
method MDGCN uses dynamic graph convolution on hyperspectral images, adapting to local regions.
result MDGCN outperforms state-of-the-art methods on benchmark datasets.
Adaptive object detection method synthesizes target domain images from source domain images.
problem Large domain gap between source and target domains in object detection.
method Cross-domain CutMix with adversarial learning.
result Higher accuracy in different domain settings compared to conventional methods.
A new framework detects novel classes in data streams.
problem Detecting novel class labels in data streams.
method Semi-supervised multi-task learning framework for co-representation learning.
result Superior performance over existing methods on real-world datasets.
Paper tackles discrimination in predictions using causal modeling.
problem Mathematical guarantee for non-discrimination in predictions.
method Causal modeling to define discrimination, bounding prediction discrimination probability.
result Discrimination in predictions can still exist even if training data is free of discrimination.
Typical cohorts in brain imaging studies are not large enough for systematic testing of all the information contained in the images. To build testable working hypotheses, investigators thus rely on analysis of previous work, sometimes formalized in a so-called meta-analysis. In brain imaging, this approach underlies th…
The study quantifies and compares aleatoric and epistemic discrimination in ML models.
problem Sources of discrimination in ML models and their impact on performance.
method Quantifying aleatoric and epistemic discrimination using statistical experiments and model accuracy.
result State-of-the-art fairness interventions are effective at removing epistemic discrimination but not aleatoric discrimination in datasets with missing values.
Deep model diagnoses psychiatric disorders from fMRI data.
problem Diagnosing psychiatric disorders from brain imaging data.
method Deep neural generative model using Bayes' rule.
result Improved diagnostic accuracy over competitive models.
Discriminator guidance improves autoregressive diffusion models for generating molecular graphs.
problem Improving the accuracy of autoregressive diffusion models for generating molecular graphs.
method Deriving ways to use a discriminator with a pretrained generative model in the discrete case, including optimal and sub-optimal scenarios.
result Using a discriminator can correct pretrained models and improve exact sampling from the data distribution.
This paper proposes a curriculum learning method for GANs using multiple discriminators.
problem Training GANs with sufficient convergence conditions and avoiding mode collapse.
method A framework for training the generator against an ensemble of discriminator networks, formalized in the full-information adversarial bandit framework.
result Our approach improves samples quality and diversity over existing baselines by effectively learning a curriculum.
Solves GAN mode collapse by assigning minibatches to multiple discriminators.
problem Mode collapse in GANs where models generate similar samples.
method Multiple discriminators, microbatching, and changing tasks.
result Promotes sample diversity in generated sets.