CNNs struggle with negative images, showing lower accuracy.
problem The limitation of CNNs in recognizing negative images.
method Evaluation of CNNs on negative images to assess their semantic understanding.
result CNNs perform poorly on negative images, suggesting lack of generalization.
This paper shows how optimizing with hard negative examples improves image retrieval.
problem Training with hard negative examples leads to poor training behavior.
method Characterize the space of triplets, derive why hard negatives fail, and offer a fix to the loss function.
result Optimizing with hard negative examples leads to more generalizable features and better image retrieval.
NegToMe uses images to guide text-based models away from unwanted visual elements.
problem Insufficient text-based adversarial guidance for complex visual concepts.
method Negative token merging (NegToMe) using visual features from reference images.
result Significantly enhances output diversity and reduces visual similarity to copyrighted content.
Improved visual representation learning with conditional negative sampling.
problem Learning strong unsupervised visual representations using contrastive learning.
method Introduce a family of mutual information estimators that sample negatives conditionally.
result Improves accuracy by 2-5% points on four standard image datasets.
Non-negative constraints improve neural network defenses.
problem Effective defenses against adversarial attacks in neural networks.
method Non-negative weight constraints applied to binary and non-binary classification problems.
result Non-negative constraints can improve resistance to adversarial attacks, especially in binary classification with asymmetric costs.
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.
Enhances contrastive learning for better representation learning on wild images.
problem Binary partition of views from same and different instances limits CL's performance.
method Doubly Contrastive Learning (CACR) with contrastive attraction and repulsion.
result CACR improves performance and robustness on wild image datasets.
Study examines how digital image alterations affect AI classification models.
problem Impact of digital alterations on image classification models.
method Evaluation of state-of-the-art machine learning models under various digital image alterations.
result Discoveries in training techniques to enhance model robustness.
Model learns image-word associations from captions using contrastive learning.
problem Phrase grounding, associating image regions to caption words.
method Optimizing word-region attention to maximize mutual information, using language model guided word substitutions for negatives.
result Model achieves 76.7% accuracy on Flickr30K Entities benchmark, a 5.7% gain from weak supervision.
The study proves curvature implications for Weil-Petersson metrics.
problem Negative-curvature criterion for Weil-Petersson metric.
method Semipositivity analysis of vector bundles and direct images.
result Negative-curvature criterion for generalized Weil-Petersson metric.
Study of negative ads on social media during U.S. midterm elections.
problem Understanding the effectiveness and mechanisms of negative advertising on social media.
method Machine learning for sentiment analysis, AI image recognition, ordinal regressions.
result Negative ads are less effective than previously thought, anger is a key mechanism.
Enhances VAEs for sharper image synthesis.
problem Blurriness in generated images from VAEs.
method Integrates a downscaled version of the original image into the VAE framework and uses it as input to the decoder.
result Improves FID score in image synthesis while maintaining similar log-likelihood performance.
Constructs a convex Finsler metric on vector bundles under specific conditions.
problem Creating a convex Finsler metric on vector bundles with positive curvature.
method Uses the negativity of direct image bundles and Minkowski inequality for norms.
result Shows how to upgrade a Kobayashi positive Finsler metric to a convex one.
Study on Lie groups with negative Ricci curvature, including open questions and a new cone.
problem Understanding Lie groups with negative Ricci curvature metrics.
method Overview and introduction of a new cone C(n) for solvable Lie algebras.
result Introduction of a new open cone C(n) that parametrizes solvable Lie algebras with negative Ricci curvature metrics.
BYOL learns useful representations without batch statistics.
problem BYOL's reliance on batch statistics for representation learning.
method Training BYOL without batch normalization and using alternative normalization schemes.
result BYOL performance comparable to vanilla BYOL without batch normalization.
Improved analysis of calcium imaging data with higher SNR and compression.
problem Challenges in analyzing large-scale calcium imaging datasets.
method Spatially-localized penalized matrix decomposition (PMD) for denoising and compression.
result Significant improvement in SNR and compression rates with minimal signal loss.
The paper reviews techniques for detecting errors in semantic segmentation models.
problem Detecting false positives and false negatives in semantic segmentation models.
method Uncertainty quantification techniques applied to semantic segmentation.
result Techniques for detecting false positives and false negatives are proposed and discussed.
VAE enhances NMF for probabilistic non-negative matrix factorisation.
problem Non-negative matrix factorisation with probabilistic coefficients.
method Design a VAE network with non-negative weights and non-negative Weibull distribution.
result Effective probabilistic NMF for generating new data and linking latent and input variables.
Counterexamples show Salter's question on Burau image is negative for n=4.
problem Conditions for a matrix to be in the Burau image of B4. method Analyzing the central quotient and using counterexamples.
result The central quotient of the Burau image group does not coincide with the central quotient of a specific subgroup of the unitary group for n=4. A new SSL method using whitening of latent-space features.
problem Efficiency and effectiveness of self-supervised representation learning.
method Proposes a new loss function based on whitening of latent-space features, avoiding the need for negatives and asymmetric networks.
result Improves efficiency and effectiveness of SSL by avoiding the need for negatives and asymmetric networks.
Authors prove a formula relating the Gaussian curvature of polyhedral vertex stars to their Gauss images.
problem Proving a formula connecting discrete Gaussian curvature to the algebraic area of Gauss images.
method Comparing winding numbers and critical point index of a normal vector to deduce the formula.
result Formula significantly limits possible shapes of Gauss images of polyhedral vertex stars.
Reduces false positives in lung nodule detection by using unlabeled data.
problem Lack of labeled data for training supervised algorithms in medical imaging.
method Uses pseudo-negative labels from unlabeled data to refine a pulmonary nodule detection network.
result False positive rate reduced from 0.4864 to 0.1266 while maintaining sensitivity.
We introduce the concept of "negative bubbles" as the mirror image of standard financial bubbles, in which positive feedback mechanisms may lead to transient accelerating price falls. To model these negative bubbles, we adapt the Johansen-Ledoit-Sornette (JLS) model of rational expectation bubbles with a hazard rate de…
The paper surveys methods to approximate non-negative matrices using lower-dimensional factors.
problem Approximating high-dimensional non-negative matrices with lower-dimensional factors.
method Alternating minimization with surrogate functionals for Tikhonov functionals.
result Developed a general framework for adding penalty terms to surrogate functionals.
The paper proposes a new method for online image decomposition using auto-encoders.
problem Building a part-based representation of image datasets for interpretation and online computation.
method Sparse, non-negative auto-encoder with deep encoder and shallow decoder for online computation.
result The method outperforms state-of-the-art online methods on MNIST and Fashion MNIST datasets.
BYOL learns image representations without negative pairs.
problem Self-supervised image representation learning.
method Two neural networks interact and learn from each other, updating the target network with a slow-moving average of the online network.
result Achieves state-of-the-art performance on ImageNet.
A famous result of Bennequin states that for any braid representative of the unknot the Bennequin number is negative. We will extend this result to all n-trivial closed n-braids. This is a class of infinitely many knots closed under taking mirror images. Our proof relies on a non-standard parametrization of the Homfly …
This research tackles ocean remote sensing data enhancement using locally-adapted convolutional models.
problem Super-resolution of irregularly-sampled ocean remote sensing images.
method Optimal interpolation as low-resolution reconstruction, locally-adapted multimodal convolutional models, and dictionary-based decompositions (PCA, sparse priors, non-negativity constraints).
result Locally-adapted parametrizations with non-negativity constraints outperform optimally-interpolated reconstructions.
Green's functions and Biot-Savart operators on curved spaces quantify linking numbers.
problem Calculating linking numbers on curved spaces.
method Constructing radial fundamental solutions for differential form Laplacian.
result Green's functions and Biot-Savart operators link curved spaces' geometry to linking numbers.
Improves domain adaptation by aligning source and target distributions and mitigating noisy labels.
problem Improving performance on target images with different acquisition conditions.
method Combines optimal transport, MixUp regularization, and robust loss for noisy labels.
result Improves domain adaptation performance on various benchmarks and real-world problems.
A Steiner deltoid maintains constant area across all boundary points of an ellipse.
problem Finding curves associated with ellipses with constant area.
method Negative Pedal Curve of the Ellipse with respect to a boundary point M.
result The Steiner deltoid has constant area over all boundary points.
One-harmonic maps from a curved surface to hyperbolic plane have specific interior properties.
problem Characterizing one-harmonic maps from curved surfaces to hyperbolic spaces.
method Using Minkowski geometry and interpreting maps as Gauss maps of convex surfaces.
result One-harmonic maps have images confined to the interior of convex hulls.
Algorithm for fast matrix factorization of large datasets.
problem Factorizing huge matrices with sparse or dense factors.
method Subsampling and iterative learning of matrix factors.
result Significant speed-ups on large datasets.
New mutual information framework improves contrastive learning for vision tasks.
problem Maximizing mutual information for better unsupervised learning representations.
method Reformulated mutual information as a lower bound, introducing new negative sampling strategies.
result Improved representations outperform previous methods in various vision tasks.
Proposes a new loss function for deep neural networks.
problem Deep neural networks lack a direct method to discriminate between correct and competing classes.
method Introduces a discriminative loss function based on negative log likelihood ratio.
result Significantly outperforms cross-entropy loss on image classification tasks.
Paper presents robust hyperspectral unmixing using correntropy maximization.
problem Robust unmixing of hyperspectral images with noisy bands.
method Maximizes correntropy criterion using ADMM for non-negativity and sparsity constraints.
result ADMM-based correntropy maximization is robust to noisy bands.
A fast k-NN classifier without negative pairs.
problem Efficient k-NN classification without negative pairs. method Ridge regression for learning dissimilarity function.
result Better k-NN classification accuracy than state-of-the-art methods. Deep learning model detects and flags artefacts in polarimetric images.
problem Artifacts in polarimetric images contaminate areas of interest.
method Convolutional Neural Network (CNN) for automatic artefact detection.
result Model achieves 98% true positive and 97% true negative rates.
Extends conformal prediction to contrastive learning for better coverage of positive samples.
problem Lack of principled guarantees on coverage in contrastive learning.
method Introduces minimum-volume covering sets with learnable constraints.
result Improves inclusion-exclusion trade-offs in positive and negative samples.
Paper improves zero-shot classification by mining hard negative pairs.
problem Zero-Shot Classification (ZSC) problems.
method Metric learning with controlled negative pairs.
result Significant improvement in performance on ZSC datasets.
New normalization technique balances positive and negative weights for faster convergence.
problem Balancing positive and negative weights for faster convergence.
method Transformation of layer weights instead of outputs, balancing positive and negative contributions.
result Balanced normalization leads to faster convergence on standard benchmarks.
Unified framework for non-negative matrices and tensors using Wasserstein loss.
problem Finding low-dimensional representations of high-dimensional datasets with non-negative constraints.
method Unified mathematical framework with a smoothed Wasserstein loss, convex dual formulation for efficient computation.
result Efficient solution for non-negative matrix and tensor factorisations with Wasserstein loss.
Solved a specific case of Salter's question on Burau representation.
problem Under what conditions are matrices in the image of the Burau representation of B3. method Algorithmically constructed a counterexample to Salter's specific question.
result The central quotient of the Burau image group is not the central quotient of a certain subgroup of the unitary group.
Detects memorization in neural networks using non-negative factorization.
problem Identifying memorization in deep neural networks.
method Measures non-linearity using non-negative factorization of activation matrices.
result High non-linearity in deep layers indicates memorization.
New methods protect malware classification networks from adversarial attacks.
problem Adversarial perturbations compromise malware classification networks.
method Training restricted networks with non-negative weight restrictions and relaxing constraints.
result Improved classifier accuracy while maintaining resistance to adversarial attacks.
By combining (i) the economic theory of rational expectation bubbles, (ii) behavioral finance on imitation and herding of investors and traders and (iii) the mathematical and statistical physics of bifurcations and phase transitions, the log-periodic power law (LPPL) model has been developed as a flexible tool to detec…
The study uses information theory to set lower bounds on model likelihoods.
problem Improving latent variable models by optimizing priors or likelihoods.
method Applying rate-distortion theory to find lower bounds on negative log likelihood.
result Rate-distortion theory can be used to optimize priors and likelihoods in latent variable models.
Proposes a new method for posterior sampling using MMD with negative distance kernel.
problem Posterior sampling and conditional generative modeling.
method Approximates joint distribution using discrete Wasserstein gradient flows of MMD with negative distance kernel.
result Establishes an error bound for posterior distributions and proves the method is a Wasserstein gradient flow.