arXiv research

Symmetric losses improve classifier robustness from corrupted labels.

This paper aims to provide a better understanding of a symmetric loss. First, we emphasize that using a symmetric loss is advantageous in the balanced error rate (BER) minimization and area under the receiver operating characteristic curve (AUC) maximization from corrupted labels. Second, we prove general theoretical p…

Symmetrizes loss functions to improve neural network robustness against noisy labels.

Deep networks can memorize random labels; symmetric loss improves this.

This paper finds ReLU restores symmetry in SCL under class imbalances.

We extend the classical theory of isothermic surfaces in conformal 3-space, due to Bour, Christoffel, Darboux, Bianchi and others, to the more general context of submanifolds of symmetric $R$-spaces with essentially no loss of integrable structure.

Optimal weight windows are symmetric rectangles centered at peak.

S-SGD adds symmetrical noise to weights to avoid sharp minima in deep learning.

We study the robustness to symmetric label noise of GNNs training procedures. By combining the nonlinear neural message-passing models (e.g. Graph Isomorphism Networks, GraphSAGE, etc.) with loss correction methods, we present a noise-tolerant approach for the graph classification task. Our experiments show that test a…

Supervised learning is an active research area, with numerous applications in diverse fields such as data analytics, computer vision, speech and audio processing, and image understanding. In most cases, the loss functions used in machine learning assume symmetric noise models, and seek to estimate the unknown function …

This paper calculates worst-case target semi-variances for uncertain losses.

We study losses for binary classification and class probability estimation and extend the understanding of them from margin losses to general composite losses which are the composition of a proper loss with a link function. We characterise when margin losses can be proper composite losses, explicitly show how to determ…

Study explores loss design for decision trees to improve robustness against noisy labels.

Study improves $H$-consistency bounds for regression analysis.

An econometric or statistical model may undergo a marginal gain if we admit a new variable to the model, and a marginal loss if we remove an existing variable from the model. Assuming equality of opportunity among all candidate variables, we derive a valuation framework by the expected marginal gain and marginal loss i…

This work shows that Gaussian is the only prior for optimal linear estimation in $L^1$ loss.

Uniqueness of 1D bi-Schrödinger flow proven from flat torus to compact space.

A new hypergraph expansion method treats vertices and hyperedges equally, improving node classification.

Study of loss functions for learning to defer, proving consistency.

Framework for systemic risk modeling using jointly exchangeable arrays.

Efficiently estimates mean of symmetric distributions without moments.

We introduce the Mutual Information Machine (MIM), a novel formulation of representation learning, using a joint distribution over the observations and latent state in an encoder/decoder framework. Our key principles are symmetry and mutual information, where symmetry encourages the encoder and decoder to learn differe…

Algorithm finds optimal regularizers for online linear optimization.

A fundamental question in deep learning concerns the role played by individual layers in a deep neural network (DNN) and the transferable properties of the data representations which they learn. To the extent that layers have clear roles, one should be able to optimize them separately using layer-wise loss functions. S…

TRS-ODENs learn dynamics with time-reversal symmetry for more efficient learning.

The SCMU algorithm computes cone factorizations for symmetric cones, improving upon existing methods.

Improved GAN training stability through tunable classification losses.

This paper tackles gauge fixing and regularity for perturbations around spherical backgrounds.

Study shows directional convergence for neural networks under spherical symmetry.

Study of historic stock returns distributions, highlighting asymmetry and outliers.

Paper proves conditions for estimating precision matrices with Laplacian constraints.

New algorithm reduces online learning regret for bounded recall games.

New method improves mean estimation for heavy-tailed data.

Boosting theory extended to handle cost-sensitive and multi-objective losses.

Extends double linear policy with time-varying weights and proves robust positive expectation.

Label smoothing improves model performance even with noisy labels.

ANTIDOTE reduces noisy labels influence during learning.

Paper tackles noisy labels in deep learning networks.

We introduce SARR for symmetric object pose estimation, improving CNN performance.

In this paper, we study how to solve resource allocation problems in ultra-reliable and low-latency communications by unsupervised deep learning, which often yield functional optimization problems with quality-of-service (QoS) constraints. We take a joint power and bandwidth allocation problem as an example, which mini…

New method estimates model risk without knowing function class.

The paper explores conditions for predicting optimization performance.

We rigorously prove statistical physics predictions for non-convex GLMs in high dimensions.

We analyze algorithms for approximating a function $f(x) = Φx$ mapping $\Re^d$ to $\Re^d$ using deep linear neural networks, i.e. that learn a function $h$ parameterized by matrices $Θ_1,...,Θ_L$ and defined by $h(x) = Θ_L Θ_{L-1} ... Θ_1 x$. We focus on algorithms that learn through gradient descent on the population …

Study detects boundaries in unlabeled noisy images without labels.

Paper explains neural collapse in neural networks using a new model.

The geometrical features of the (non-convex) loss landscape of neural network models are crucial in ensuring successful optimization and, most importantly, the capability to generalize well. While minimizers' flatness consistently correlates with good generalization, there has been little rigorous work in exploring the…

We address the problem of detecting changes in multivariate datastreams, and we investigate the intrinsic difficulty that change-detection methods have to face when the data dimension scales. In particular, we consider a general approach where changes are detected by comparing the distribution of the log-likelihood of …