Feature noise causes loss discrepancies across groups even with equal data.
problem Loss discrepancies observed in learning procedures across different groups.
method Characterized the effect of feature noise on loss discrepancy in linear regression.
result Feature noise leads to loss discrepancy even when groups have equal data.
Calibrating classifiers reduces grouping loss using sufficiency criteria.
problem Grouping loss in probabilistic classifier calibration is often overlooked.
method Revisited Langford & Zadrozny's probing reduction approach and introduced Brier curves.
result The probing reduction approach reduces grouping loss and supports sufficient calibration.
New method estimates grouping loss in neural networks to improve confidence scores.
problem Improving confidence scores in neural networks to reflect true posterior probabilities.
method Proposed an estimator to approximate the grouping loss.
result Modern neural networks exhibit grouping loss, especially in distribution shifts.
Though machine learning algorithms excel at minimizing the average loss over a population, this might lead to large discrepancies between the losses across groups within the population. To capture this inequality, we introduce and study a notion we call maximum weighted loss discrepancy (MWLD), the maximum (weighted) d…
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 study a family of sparse estimators defined as minimizers of some empirical Lipschitz loss function -- which include the hinge loss, the logistic loss and the quantile regression loss -- with a convex, sparse or group-sparse regularization. In particular, we consider the L1 norm on the coefficients, its sorted Slope…
Loss minimization leads to multicalibration for neural networks.
problem Ensuring fairness in predictions across multiple protected groups.
method Minimizing squared loss over neural networks of size n.
result Minimizing loss over neural nets of size n implies multicalibration for most values of n.
Improved neural networks by combining group DRO with regularization.
problem Overparameterized neural networks can fail on atypical groups due to spurious correlations.
method Coupling distributionally robust optimization (DRO) with increased regularization.
result Significant improvements in worst-case group accuracy, maintaining high average accuracy.
Machine learning models (e.g., speech recognizers) are usually trained to minimize average loss, which results in representation disparity---minority groups (e.g., non-native speakers) contribute less to the training objective and thus tend to suffer higher loss. Worse, as model accuracy affects user retention, a minor…
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.
Many practical applications such as gene expression analysis, multi-task learning, image recognition, signal processing, and medical data analysis pursue a sparse solution for the feature selection purpose and particularly favor the nonzeros \emph{evenly} distributed in different groups. The exclusive sparsity norm has…
Most systems and learning algorithms optimize average performance or average loss -- one reason being computational complexity. However, many objectives of practical interest are more complex than simply average loss. This arises, for example, when balancing performance or loss with fairness across people. We prove tha…
Methodology measures financial impacts using existing credit loss infrastructure.
problem Measuring the impact of financial scenarios on expected credit losses.
method Captures scenario effects through changes in default probabilities; uses existing provisioning infrastructure.
result Methodology validated through standardized climate scenario exercise in Canada and Quebec.
New loss function improves classification for imbalanced and sensitive groups.
problem Optimizing metrics like balanced error and equal opportunity in imbalanced and sensitive classification.
method Developed a principled vector-scaling (VS) loss function that addresses multiplicative adjustments necessary at terminal training phase.
result The VS loss function improves minority class performance and generalizes to different types of imbalances.
This research tackles group fairness in predictive process monitoring by ensuring predictions are independent of sensitive group membership.
problem Predictive models using biased historical data can perpetuate unfair behavior in new cases.
method Investigates independence through metrics like ΔDP and a composite loss function balancing predictive performance and fairness.
result Proposes and validates a composite loss function for training models that balance fairness and performance.
New method reduces privacy impact on model accuracy for underrepresented groups.
problem Privacy mechanisms disproportionately affect underrepresented groups in machine learning models.
method Proposes DPSGD-F, a modified DPSGD that adjusts group contributions based on clipping bias.
result DPSGD-F removes disparate impact of differential privacy on model accuracy for protected groups.
New method finds sparse groups of input variables for neural networks.
problem Finding optimal groups of input variables for neural networks.
method Developed a new loss function and optimization algorithm for multi-layer non-linear neural networks to achieve group sparsity.
result Achieved group sparsity in three real-world datasets, improving model performance and excluding a significant number of variables.
Paper introduces a new robust method for estimating Pareto tail index from grouped data.
problem Limited robust methods for estimating Pareto tail index from grouped data.
method Method of Truncated Moments (MTuM)
result Inferential justification and validation of MTuM through simulation study.
New method approximates curvature from symmetries in deep networks.
problem Hard to approximate curvature in large deep networks.
method Analytically averaging over group actions that leave the loss invariant to construct structured Hessian approximations.
result Structured Hessian approximations from single gradients can be estimated, stored, and inverted.
Model predicts internal fraud in retail banking is cyclical and influenced by corruption.
problem Predicting and mitigating internal fraud losses in retail banking.
method Developed a dynamic model considering internal factors and macroeconomic indicators.
result Internal fraud losses are pro-cyclical and positively affected by corruption perceptions.
This paper improves operational risk modeling by selecting better loss severity distributions.
problem Inconsistent regulatory capital calculations due to changing loss severity distribution families.
method Presented truncation probability estimates and a consistent quantile scoring function for selection criteria. Also, recommended collecting loss frequencies below the minimum reporting threshold.
result More stable regulatory capital calculations through better selection of loss severity distributions.
New risk control method for non-monotonic losses in complex parameters.
problem Controlling risk for non-monotonic losses with multidimensional parameters.
method Stability-based guarantees for generic algorithms applied to non-monotonic losses.
result Guarantees depend on algorithm stability, with looser guarantees for unstable algorithms.
This work classifies SOC loss functions based on their gradient properties.
problem Optimizing noisy systems in stochastic optimal control.
method Grouping loss functions into classes with the same gradient expectation.
result Different loss functions have the same optimization landscape but differ in gradient variance.
The paper decomposes probabilistic scores into reliability, uncertainty, and information loss.
problem Understanding the reliability and uncertainty of probabilistic predictions.
method Developed decomposition identities for proper losses, quantifying reliability, residual uncertainty, and information gain.
result A three-term identity for classification scores, revealing miscalibration, grouping term, and feature-level uncertainty.
The paper analyzes and proposes methods for privately sharing individual privacy losses using per-instance differential privacy.
problem The standard differential privacy framework provides a worst-case bound that may not accurately reflect individual privacy losses.
method The paper analyzes per-instance differential privacy and proposes methods to privately and accurately publish per-instance privacy losses.
result The methods privately and accurately publish per-instance differential privacy losses with minimal additional privacy cost.
Current methods to interpret deep learning models by generating saliency maps generally rely on two key assumptions. First, they use first-order approximations of the loss function neglecting higher-order terms such as the loss curvatures. Second, they evaluate each feature's importance in isolation, ignoring their int…
We present a new machine learning approach to estimate personalized treatment effects in the classical potential outcomes framework with binary outcomes. To overcome the problem that both treatment and control outcomes for the same unit are required for supervised learning, we propose surrogate loss functions that inco…
GMLP learns feature groups for tabular data without known structure.
problem Deep learning for tabular data with unknown feature interactions.
method Group-wise operations and sparse feature grouping matrix learned through temperature annealing softmax.
result GMLP achieves state-of-the-art classification performance on various datasets.
Proposes a method to ensure low losses across all subpopulations in large datasets.
problem Standard practice of minimizing average loss fails to guarantee low losses across all subpopulations in heterogeneous datasets.
method Convex procedure that controls worst-case performance over all subpopulations of a given size with finite-sample convergence guarantees.
result Empirically, the worst-case procedure learns models that do well against unseen subpopulations.
In this paper, we propose an auto-encoder based generative neural network model whose encoder compresses the inputs into vectors in the tangent space of a special Lie group manifold: upper triangular positive definite affine transform matrices (UTDATs). UTDATs are representations of Gaussian distributions and can strai…
New guarantees for Group LASSO in sparse convex optimization.
problem Sparse convex optimization with vector-valued features.
method Group LASSO regularization and analysis of gradient norms.
result Group LASSO selects the same features as Orthogonal Matching Pursuit.
ABM automates feature engineering and variable selection for loss-based models.
problem Improving model performance through better feature engineering and variable selection.
method ABM uses group and fused lasso regularization to automatically select cutting points and variables.
result ABM integrates feature engineering, variable selection, and model training.
Study achieves fairness without demographic info, improving regression tasks.
problem Achieve fairness in models without prior demographic info.
method VFair method to minimize training loss variance, dynamic update approach.
result Regression tasks can achieve significant fairness improvement without prior demographics.
It has been shown that dimension reduction methods such as PCA may be inherently prone to unfairness and treat data from different sensitive groups such as race, color, sex, etc., unfairly. In pursuit of fairness-enhancing dimensionality reduction, using the notion of Pareto optimality, we propose an adaptive first-ord…
New machine learning paradigm ignores loss function until action time.
problem Learning with unknown loss functions.
method Introduces omnipredictors for any loss function.
result Extracts predictive power from any class, ignoring loss function.
The study quantifies decision-making risks from suboptimal classifiers and proposes methods to reduce these risks.
problem Excess risk in decision-making from suboptimal probabilistic classifiers.
method Analytical expressions and upper/lower bounds for excess risk, calibration curve estimation, grouping loss estimator.
result Identifies regimes where recalibration alone or post-training is more effective.
We study the impact of contagion in a network of firms facing credit risk. We describe an intensity based model where the homogeneity assumption is broken by introducing a random environment that makes it possible to take into account the idiosyncratic characteristics of the firms. We shall see that our model goes behi…
Paper analyzes trade-offs in top-k classification accuracies and proposes a new loss function.
problem CE loss does not always optimize top-k prediction, especially with complex data.
method Introduces a novel top-k transition loss to improve top-k accuracy.
result Our loss function improves top-k accuracy, especially for k > 10.
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.
A significant advance in accelerating neural network training has been the development of normalization methods, permitting the training of deep models both faster and with better accuracy. These advances come with practical challenges: for instance, batch normalization ties the prediction of individual examples with o…
We study channel number reduction in combination with weight binarization (1-bit weight precision) to trim a convolutional neural network for a keyword spotting (classification) task. We adopt a group-wise splitting method based on the group Lasso penalty to achieve over 50% channel sparsity while maintaining the netwo…
Discover conservation laws from trajectories using a neural network.
problem Finding invariants and conservation laws from large-scale data without prior knowledge.
method ConservNet, a neural network trained with noise-variance loss to discover hidden invariants in grouped multi-dimensional observables.
result Successfully discovers underlying invariants from simulated and real-world systems.
We introduce a representation theory for risk operations on locally compact groups in a partition of unity on a topological manifold for Markowitz-Tversky-Kahneman (MTK) reference points. We identify (1) risk torsion induced by the flip rate for risk averse and risk seeking behaviour, and (2) a structure constant or co…
FedCBO solves clustered federated learning by optimizing groups of users without knowing their structure.
problem Training models for multiple users with privacy and communication constraints, especially in clustered settings.
method FedCBO uses a particle system approach inspired by consensus-based optimization to train models for each user group.
result FedCBO outperforms other methods in training models for clustered federated learning.
Background: Fluctuating hearing loss is characteristic of Meniere's Disease (MD) during acute episodes. However, no reliable audiometric hallmarks are available for counselling the hearing recovery possibility. Aims/Objectives: To find parameters for predicting MD hearing outcomes. Material and Methods: We applied mach…
Study shows sudden loss of balance in stock market networks after 2011, reducing predictability.
problem Reduced predictability in stock markets due to structural changes.
method Rank correlations and weighted signed networks to analyze interconnectivity and balance.
result Sudden loss of balance in stock market networks after 2011, leading to decreased predictability.
A new subdivision scheme for Heisenberg group values with central smoothness loss.
problem Regularity of limit curves in Heisenberg group-valued subdivision schemes.
method Interpolatory subdivision scheme with central correction based on group law.
result Central part of limit curve converges to a continuous limit with logarithmic modulus of continuity.
Significant progress has been made using fMRI to characterize the brain changes that occur in ASD, a complex neuro-developmental disorder. However, due to the high dimensionality and low signal-to-noise ratio of fMRI, embedding informative and robust brain regional fMRI representations for both graph-level classificati…