Improved CNN for HCCR with new loss function and ranking method.
problem Loss of inter-class information in traditional CNN models for HCCR.
method Combining cross entropy with a new similarity ranking function (Average variance similarity) as loss function.
result New loss function (SoftMax cross entropy with Average variance similarity) achieves highest accuracy in HCCR.
Introduces CHL, a new loss function for continuous similarity learning.
problem Binary similarity learning limitations.
method CHL is a novel loss function that generalizes histogram loss to continuous similarities.
result CHL solves a wider range of tasks including similarity learning, representation learning, and data visualization.
Exploits class similarity for better machine learning models with confidence labels and projective loss functions.
problem Poor model performance due to confusing similar classes.
method Exploits class similarity with confidence labels and projective loss functions.
result Improved model performance on noisy labels.
Improved similarity search in embeddings using InfoNCE loss.
problem Improving similarity search in embedding models trained by contrastive learning.
method Introduced a new continuity bound for InfoNCE loss via Gâteaux differentiation, preserving the averaging effect of negative samples.
result Demonstrated that the averaging effect of k negative samples in InfoNCE loss carries over to stabilisation of generalisation error as k grows. New loss function and training scheme improve binary hash codes for better similarity search.
problem Improving binary hash codes for better similarity search tasks.
method Log likelihood loss on Hamming distance target, novel training scheme, multi-indexing.
result Significant improvements in MAP (84%) and query cost reduction for ImageNet and SIFT 1M.
Task loss matching misrepresents similarity between neural network layers.
problem Measuring similarity between neural network layers using task loss matching.
method Task loss matching vs. direct matching; comparison with CCA and CKA.
result Direct matching provides a better similarity index than task loss matching.
A new loss function improves deep learning performance without class separation constraints.
problem Training deep learning architectures for classification.
method Minimizing smoothness of label signals on similarity graphs.
result The proposed loss function leads to similar classification performance as cross-entropy, with added robustness.
Extends graph similarity theory to improve MPNNs' generalization abilities.
problem Understanding MPNNs' generalization beyond training data.
method Extends graph similarity theory, assesses graph structure, aggregation, and loss functions.
result Improves understanding of MPNNs' generalization properties.
A new loss function for VAEs improves image quality and efficiency.
problem Training VAEs to generate realistic images requires a loss function that reflects human perception.
method Based on Watson's perceptual model, the loss function computes a weighted distance in frequency space, accounts for luminance and contrast masking, and is extended to color images.
result VAEs trained with the new loss function generated high-quality, less blurry images with fewer artifacts and less computational resources.
This paper tackles noise in raw datasets to improve representation learning efficiency.
problem Noise in real-world datasets degrades representation learning quality.
method Proposes denoising Cosine-Similarity (dCS) loss to learn robust representations.
result Empirical results show the dCS loss outperforms baseline objective functions.
Inverse depth scaling found in LLMs due to similar layers averaging error.
problem Understanding how depth affects loss in large language models.
method Analysis of LLMs and toy residual networks.
result Loss scales inversely proportional to depth in LLMs.
UMAP's true loss function differs from what was previously thought, focusing on nearest neighbor graph similarities.
problem Understanding the true effectiveness and mechanism of UMAP for high-dimensional data visualization.
method Deriving UMAP's effective loss function in closed form and analyzing its optimization scheme.
result UMAP aims to reproduce similarities encoded in the nearest neighbor graph, not the full high-dimensional similarities.
Proposes Group Loss for deep metric learning to improve clustering and image retrieval.
problem Improving deep metric learning for better clustering and image retrieval.
method Group Loss based on label-propagation method enforcing embedding similarity across all samples of a group.
result Shows state-of-the-art results on clustering and image retrieval on several datasets.
We propose sparsemax, a new activation function similar to the traditional softmax, but able to output sparse probabilities. After deriving its properties, we show how its Jacobian can be efficiently computed, enabling its use in a network trained with backpropagation. Then, we propose a new smooth and convex loss func…
Paper reduces Hausdorff Distance in medical image segmentation.
problem Reduction of Hausdorff Distance in medical image segmentation.
method Three novel loss functions for training CNNs to estimate and reduce HD.
result Approximately 18-45% reduction in HD without degrading other performance metrics.
SimLoss improves classification by considering class similarities.
problem Equal punishment of all misclassifications in CCE.
method Integrates class similarities into Categorical Cross Entropy using matrices constructed from task-specific knowledge.
result Significant improvements over CCE on Age Estimation and Image Classification.
New loss function optimization improves training speed and accuracy.
problem Optimizing neural network performance through loss functions.
method Genetic Loss-function Optimization (GLO) using genetic programming and CMA-ES.
result GLO loss functions lead to better performance with fewer training steps.
Generative Adversarial Networks create realistic brain MRI images.
problem Lack of sufficient training data for medical research.
method Experimented three GAN architectures with different loss functions and hyperparameter tuning.
result Achieved high-quality, realistic synthetic MRIs with gradient penalty and mini-batch similarity layer.
Cosine similarity can force points to grow in magnitude, causing convergence issues.
problem Cosine similarity loss can lead to convergence issues in deep learning.
method Analyzing under-explored settings and proposing cut-initialization.
result Cosine similarity optimization forces points to grow in magnitude, leading to convergence issues.
Stochastic Gradient Descent (SGD) is a central tool in machine learning. We prove that SGD converges to zero loss, even with a fixed (non-vanishing) learning rate - in the special case of homogeneous linear classifiers with smooth monotone loss functions, optimized on linearly separable data. Previous works assumed eit…
New loss function for learning sparse representations from data.
problem Emergence of sparse representations in neural networks.
method Analysis of input data distribution and regularization.
result Introduction of a new loss function for sparse regularization.
Study compares metric learning loss functions for speaker verification.
problem Comparing metric learning loss functions for end-to-end speaker verification.
method Cross entropy loss, cosine loss, angular margin loss, center loss, contrastive loss, triplet loss.
result Additive angular margin loss outperforms other loss functions.
It is widely conjectured that the reason that training algorithms for neural networks are successful because all local minima lead to similar performance, for example, see (LeCun et al., 2015, Choromanska et al., 2015, Dauphin et al., 2014). Performance is typically measured in terms of two metrics: training performanc…
COREL learns latent representations that naturally cluster, outperforming CCE.
problem Training neural networks to learn useful latent representations.
method Attractive-Repulsive Loss Framework for Clustering-Oriented Representation Learning (COREL).
result COREL variants outperform CCE in various classification tasks.
Proposes a new loss function for causal machine-learning.
problem Lack of a well-defined loss function for causal machine-learning.
method Introduces a novel loss function equal to MSE in a standard regression problem.
result Demonstrates that gradient descent can be directly applied to this loss function.
New robust metric learning method improves performance in noisy data.
problem Label noise and outliers in training data degrade metric learning performance.
method Rescaled Hinge loss function and HQ algorithm.
result New method outperforms state-of-the-art methods in noisy data.
Paper analyzes statistical properties of log-cosh loss function.
problem No statistical analysis of log-cosh loss function in literature.
method Presented statistical properties of log-cosh loss function, compared to Cauchy distribution, and examined various statistical procedures.
result Characterized statistical properties of log-cosh loss function, including distribution, likelihood function, and Fisher information.
As a highlighting research topic in the multimedia area, cross-media retrieval aims to capture the complex correlations among multiple media types. Learning better shared representation and distance metric for multimedia data is important to boost the cross-media retrieval. Motivated by the strong ability of deep neura…
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.
Adapts auxiliary losses using gradient similarity to improve neural network performance.
problem Statistical inefficiency in neural networks and difficulty in selecting helpful auxiliary tasks.
method Uses cosine similarity between gradients of tasks to adaptively weight auxiliary losses.
result Guaranteed convergence to critical points of the main task and practical usefulness across domains.
Investment horizon approach has been used to analyze indexes of Polish stock market.Optimal time horizon for each return value is evaluated by fitting appropriate function form of the distribution. Strong asymmetry of gain-loss curves is observed for WIG index, whereas gain and loss curves look similar for WIG20 and fo…
The paper examines statistical properties of IL and LVR in automated market makers.
problem Assessing the performance of automated market makers and their profitability.
method Analysis of random walk properties and statistical integral combined with CFMM mechanics.
result IL and LVR have identical expectation values but different distribution functions for Brownian motion.
The paper analyzes the generalization of deep neural networks for metric and similarity learning.
problem Lack of rigorous understanding of generalization performance in metric and similarity learning.
method Derive explicit form of true metric, construct structured deep ReLU neural network, establish excess risk bounds.
result Explicit excess risk bounds for metric and similarity learning are derived.
Paper proposes a new softmax loss for better performance in Positive and Unlabeled data tasks.
problem Current softmax losses and sampling schemes have drawbacks in Positive and Unlabeled learning.
method Proposes Relaxed Softmax (RS) loss and a new negative sampling scheme.
result New training objective drives uplifts in performance on textual and recommendation datasets.
ICE proposes a new loss function for deep metric learning.
problem Deep metric learning from instance-level matching distribution.
method Instance Cross Entropy (ICE) loss function.
result ICE outperforms existing methods on real-world benchmarks.
Proposes a new loss function for deep learning image segmentation.
problem Efficiency and reliance on labeled data in deep learning image segmentation.
method Mumford-Shah functional adapted for deep learning.
result Improves efficiency and effectiveness of deep learning for image segmentation.
New method learns distances and similarities robustly from noisy data.
problem Scalability and robustness in metric learning for large datasets.
method Robust online Distance-Similarity learning with Rescaled hinge loss.
result Significantly outperforms state-of-the-art methods in noisy data.
The study optimizes bounds for comparing training and population loss.
problem Optimizing bounds for comparing training and population loss.
method Derives generic information-theoretic and PAC-Bayesian generalization bounds using convex comparator functions.
result The tightest possible bound is obtained with the comparator being the convex conjugate of the CGF of the bounding distribution.
Vector embedding is a foundational building block of many deep learning models, especially in natural language processing. In this paper, we present a theoretical framework for understanding the effect of dimensionality on vector embeddings. We observe that the distributional hypothesis, a governing principle of statis…
We study the error landscape of deep linear and nonlinear neural networks with the squared error loss. Minimizing the loss of a deep linear neural network is a nonconvex problem, and despite recent progress, our understanding of this loss surface is still incomplete. For deep linear networks, we present necessary and s…
DNN-based cross-modal retrieval has become a research hotspot, by which users can search results across various modalities like image and text. However, existing methods mainly focus on the pairwise correlation and reconstruction error of labeled data. They ignore the semantically similar and dissimilar constraints bet…
A new framework for clustering with uncertainty quantification.
problem Lack of uncertainty quantification in clustering methods.
method Generalized Bayes framework using Gibbs posteriors and loss functions.
result Efficient algorithms for clustering and uncertainty quantification.
Structured entropy improves classification performance on structured targets.
problem Cross-entropy loss fails to account for target variable structure.
method Proposes structured entropy, a generalization of entropy using random partitions.
result Structured cross-entropy loss yields better results on classification problems with known structure.
Paper estimates differences in multi-attribute Gaussian graphical models using non-convex penalties.
problem Estimating differences in multi-attribute Gaussian graphical models with similar structure.
method Penalized D-trace loss function with non-convex (log-sum and SCAD) penalties, proximal gradient descent methods.
result Theoretical analysis and numerical examples support consistency in support recovery and estimation.
Method generates visual explanations for similarity models without classification.
problem Lack of visual explanations for similarity models trained without classification loss.
method Gradient-based visual attention using learned feature embeddings.
result Attention maps improve model performance and can be used as constraints.
New hashing method improves document retrieval precision.
problem Efficiently retrieving similar documents from large text databases.
method Pairwise supervised hashing with Bernoulli VAE and unbiased gradient estimator.
result Superior performance compared to existing methods.
Much of the focus in machine learning research is placed in creating new architectures and optimization methods, but the overall loss function is seldom questioned. This paper interprets machine learning from a multi-objective optimization perspective, showing the limitations of the default linear combination of loss f…
New insights on pruning deep networks by preserving function locality.
problem Designing effective pruning methods for deep neural networks.
method Revisited loss modeling using first and second order Taylor expansions, emphasizing locality.
result Both first and second order Taylor expansions can achieve similar performance in pruning.