Tunable sensitivity helps neural networks learn from hard examples.
problem Improving neural network learning from hard examples.
method A new generalized cross-entropy gradient step with tunable sensitivity.
result Test prediction error is lower with tunable sensitivity.
This paper enhances the Random Survival Forest model for better predictive maintenance.
problem Improving time-to-failure estimation in predictive maintenance with Random Survival Forest.
method A three-level framework for quantifying hyperparameter tunability, including model-level and hyperparameter-level metrics.
result Hyperparameter tuning significantly improves Random Survival Forest model performance in predictive maintenance.
A new GAN model α α α -GAN with tunable loss function addresses gradient vanishing and mode collapse issues.
problem Addressing vanishing gradients and mode collapse in GANs.
method Introduced a tunable GAN α α α -GAN using a supervised α α α -loss function. result Holistic understanding of α α α -GAN related to Arimoto divergence and convergence properties. A new loss function α \alpha α -loss improves classification robustness and calibration.
problem Improving classification robustness and calibration in machine learning.
method Introduces a tunable loss function α \alpha α -loss, parameterized by α \alpha α , and analyzes its theoretical and practical properties. result The α \alpha α -loss function can improve model robustness to label flips and sensitivity to imbalanced classes. Introduces tunable basis functions for Gaussian processes.
problem Reduces computational complexity in Gaussian process approximations.
method Introduces tunable, local, and bounded basis functions for kernel approximation.
result Demonstrates superior performance compared to state-of-the-art methods, especially with poorly chosen kernel functions.
Tunable GMM kernels improve on original GMM in various classification tasks.
problem Improving the efficiency and performance of GMM kernels.
method Developed three tunable GMM kernels: eGMM, pGMM, and epGMM.
result Tunable GMM kernels typically improve over the original GMM kernel on 60 datasets.
Proposes tunable GMM kernels for classification tasks.
problem Lack of competitive performance of GMM kernels compared to tree methods on deep learning datasets.
method Introduces tunable GMM kernels with added parameters and combines basic kernels for improved performance.
result Tunable GMM kernels can produce good results for various classification tasks.
Quantum memristors created in photonic platforms with real-time control.
problem Creating quantum memristors for integrated quantum photonics.
method Designing a tunable beam splitter with real-time control, using weak measurements and classical feedback.
result Demonstrated that the tunable beam splitter behaves as a quantum memristor.
MLtuner automates tuning machine learning parameters for better performance.
problem Manual tuning of machine learning parameters is error-prone and requires domain knowledge.
method Snapshotting, branching, and optimization-guided online trial-and-error.
result MLtuner finds and re-tunes parameters robustly and efficiently for various ML applications.
The paper formalizes hyperparameter tuning and benchmarks six algorithms.
problem Optimizing machine learning hyperparameters.
method Formalized problem, defined defaults, conducted benchmarking study.
result Default values and tunability measures for hyperparameters.
A framework identifies worst-case decision points in safety-critical scenarios, improving risk assessment by 10 hours.
problem Identifying worst-case outcomes in safety-critical decision-making under uncertainty.
method Explicitly estimating distributions of expected return to identify dead-ends, tuning based on risk tolerance.
result Significantly improves risk assessment, providing indications 10 hours earlier and increasing detection by 20%.
Develops a method for constructing KBs with tunable precision for subjective and factual attributes.
problem Complexity in measuring subjective attributes complicates precision estimation in KBs.
method Probabilistically models user consensus with respect to each entity-attribute pair, using neural networks to fit the model.
result Learned models can successfully control KB's precision and outperform baselines in attribute prediction.
New method reduces synchronization costs in parallel optimization.
problem Inhibiting scalability of optimization methods due to communication costs.
method Extending Communication-Avoiding Krylov subspace methods to first-order, block coordinate descent methods.
result SA-variants attain large speedups of up to 5.1x on a supercomputer.
Approximate dynamic programming (ADP) has proven itself in a wide range of applications spanning large-scale transportation problems, health care, revenue management, and energy systems. The design of effective ADP algorithms has many dimensions, but one crucial factor is the stepsize rule used to update a value functi…
A new framework for risk-aware multi-armed bandits tackles volatile environments.
problem Volatility in healthcare and finance makes naive reward maximization unreliable.
method Risk-aware strategies with adaptive risk measures and change-point detection.
result Finite-time theoretical guarantees and asymptotic regret bound of order i l d e O ( K T T ) ilde O(\sqrt{K_T T}) i l d e O ( K T T ) . Improved GAN training stability through tunable classification losses.
problem Training instabilities in GANs.
method Reformulated GAN value function using class probability estimation (CPE) losses, defined ( α D , α G ) (α_D,α_G) ( α D , α G ) -GANs. result Tuning ( α D , α G ) (α_D,α_G) ( α D , α G ) can alleviate training instabilities. A new loss function α α α -loss bridges log-loss and 0 0 0 - 1 1 1 loss for binary classification.
problem Improving binary classification performance using a tunable loss function.
method Introducing α α α -loss, proving its margin-based form and classification-calibration, and providing an upper bound on empirical risk. result Empirical and expected risk difference upper bound for logistic regression-based classification.
The paper introduces extremal perturbations for better attribution analysis in deep networks.
problem Identifying input parts responsible for model outputs.
method Extremal perturbations, smooth masks, and technical innovations for computation.
result Demonstrates excellent sensitivity to spatial properties of deep neural networks.
PINNs struggle with increasingly complex ODEs, especially when parameters control their complexity.
problem Evaluating physics-informed neural networks on complex coupled ODEs.
method Tuned benchmarks of partial differential equations and harmonic oscillators; varying network architecture and training method.
result PINNs fail to solve complex ODEs, revealing issues like insufficient capacity, poor conditioning, and high local curvature.
The paper studies practical estimation and interpretation of Rényi transfer entropy.
problem Challenges in accurately estimating and interpreting Rényi transfer entropy.
method Systematic study of k-nearest neighbor estimator for Rényi entropy and transfer entropy.
result Effective estimates of effective Rényi transfer entropy can accurately capture directional information flow.
GAM generates global explanations of neural networks by mapping prediction landscapes.
problem Lack of interpretability in neural networks.
method GAM (Global Explanations Mapping) method that explains neural network predictions across subpopulations.
result GAM's global explanations match feature weights of interpretable models and are intuitive to practitioners.
This paper uses XAI techniques to explain meta-learning models.
problem Lack of understanding how meta-features contribute to model performance.
method XAI techniques applied to explain black-box surrogate models.
result Improved understanding of meta-features' importance and effect.
TSSM splits neural networks for parallel training with minimal accuracy loss.
problem Accuracy degradation in parallel training of deep neural networks.
method TSSM reformulates alternating minimization to achieve parallelism with minimal accuracy loss.
result TSSM achieves significant speedup without accuracy loss on multiple datasets.
A comparative study of 7 clustering algorithms in R.
problem No consensus on which clustering methods are best for a given dataset.
method Systematic comparison of 7 clustering methods on artificial datasets with various properties.
result Spectral clustering usually outperforms other methods with default configurations.
Enhances multi-modular models by directing information flow between components.
problem Improving predictive performance in multi-modular models with misspecification.
method Introduces Semi-Modular Inference (SMI) with an influence parameter to control information flow between modules.
result SMI allows for tunable and directed information flow, improving prediction in some settings.
Dual-objective GANs reduce training instabilities with tunable α-loss parameters.
problem Training instabilities in Generative Adversarial Networks (GANs).
method Introduce ( α D , α G ) (α_D,α_G) ( α D , α G ) -GANs with dual objectives modeled using α α α -loss. result Upper bounds on estimation error show improved performance under certain conditions.
EigenVI uses orthogonal function expansions for efficient variational inference.
problem Efficiently approximate complex distributions in variational inference.
method EigenVI constructs variational approximations using orthogonal function expansions, minimizing Fisher divergence.
result EigenVI provides more accurate approximations than existing methods for Gaussian BBVI.
Algorithm learns optimal parameters from infinite space for computational resource optimization.
problem Finding nearly-optimal parameters from an infinite space of tunable parameters.
method Learn a finite set of promising parameters from an infinite set using a data-independent discretization approach.
result Algorithm can help compile a configuration portfolio or select input to a configuration algorithm for finite parameter spaces.
EUNNs improve RNN performance and efficiency.
problem Gradient explosion/vanishing and long-term correlations in RNNs.
method Unitary matrices with tunable representation capacity and O ( 1 ) \mathcal{O}(1) O ( 1 ) computational complexity. result EUNNs significantly outperform other RNNs and LSTMs in performance and training speed.
TG-PSM morphs encrypted traffic to reduce traffic analysis accuracy.
problem Traffic analysis on encrypted traffic reveals sensitive information.
method Clusters websites, finds target sites, morphs traffic using greedy algorithm.
result TG-PSM reduces attacker's accuracy significantly with low overhead.
Quantum neural networks need both data-dependent and trainable unitaries for effective geometric deformation.
problem Quantum neural networks lack the geometric flexibility of classical networks due to limitations in state reachability.
method Viewing quantum states as embedded manifolds, we analyze infinitesimal unitary actions and introduce the CLA maps and aCLS criterion.
result Geometric flexibility in quantum neural networks requires a joint dependence on data and trainable weights.
ARS visualization improves t-SNE dynamics with tunable attraction and repulsion.
problem Improve data visualization techniques for complex data sets.
method ARS framework based on t-SNE dynamics with normalized interactions and tunable kernels.
result ARS visualization provides better control over cluster tightness and spacing.
A new method generates MAX-2-SAT instances with adjustable hardness.
problem Creating hard MAX-SAT instances for evaluating MAX-SAT solvers.
method Inspired by frustrated-loop algorithm, extends to bipartite couplings, tuning hardness through frustration index.
result Generated instances can be tuned through a central parameter (frustration index), showing double phase transition behavior.
New AI tools generate and vary dance choreographies.
problem Creating original and customizable dance choreographies.
method Recurrent neural networks and autoencoders trained on movement data.
result Generated and varied dance sequences using machine learning.
Proposes a Dual Focal Loss method to improve semantic segmentation on unbalanced datasets.
problem Improving performance on unbalanced datasets in semantic segmentation.
method Adaptive Class Weight based Dual Focal Loss (DFL) function.
result DFL improves convergence and reduces class imbalance effect.
A new node embedding method that adapts to graph structure.
problem Scalable node embedding for large graphs.
method Adaptive node similarity matrix for multilength paths.
result Superior performance in node classification, link prediction, and clustering.
Kernel ridgeless regression with random features shows good generalization without explicit regularization.
problem Generalization of kernel ridgeless regression without explicit regularization.
method Investigation of ridgeless regression with random features and stochastic gradient descent, exploring the effect of random features error and spectral density optimization.
result Random features error exhibits the double-descent curve, leading to improved generalization.
DiffGBM improves tree-based models for tabular regression by making their design defaults explicit and tunable.
problem Improving tree-based models for tabular regression without neural density estimators.
method Explicitly exposes and makes tunable the design defaults of tree-based models, using a Gaussian-path flow-matching trainer and a score-flex space.
result Selected configurations of DiffGBM outperform the published baseline on all eleven tabular benchmarks, with the best aggregate CRPS skill.
Paper defines hyperparameter importance for efficient tuning.
problem Computational inefficiency in tuning all hyperparameters.
method Defines hyperparameter importance via subsampling procedures.
result Proposed importance consistent with full data under weak conditions.
Margin maximization in the hard-margin sense, proposed as feature elimination criterion by the MFE-LO method, is combined here with data radius utilization to further aim to lower generalization error, as several published bounds and bound-related formulations pertaining to lowering misclassification risk (or error) pe…
Artificial neural networks estimate model parameters from observations, reducing model errors.
problem Estimating parameters of convection-permitting models from observations.
method Training Bayesian neural networks and point estimate neural networks on atmospheric state observations.
result Artificial neural networks can estimate model parameters and their statistics.
We create synthetic Morse code datasets for machine learning.
problem Creating challenging datasets for neural networks.
method Algorithm to generate synthetic Morse code datasets of varying difficulty.
result Network performance is affected by noise and feature set expansion.
We develop a scalable method for Bayesian neural networks with stochastic differential equations.
problem Uncertainty quantification in deep neural networks.
method Gradient-based stochastic variational inference in continuous-depth Bayesian neural networks.
result Gradient estimator with zero variance as the approximation improves.
In the spirit of behavioral finance, we study the process of opinion formation among investors using a variant of the 2D Voter Model with a tunable social temperature. Further, a feedback acting on the temperature is introduced, such that social temperature reacts to market imbalances and thus becomes time dependent. I…
We analyze the optimization landscape of α-loss in logistic models.
problem Optimization landscape of α-loss in logistic models.
method Tools from strictly-locally-quasi-convex functions and geometric techniques.
result Evolution of optimization landscape with respect to α.
Prevents sensitive data generation in diffusion models using labeled and unlabeled data.
problem Generating sensitive data in diffusion models using unlabeled data.
method Positive-Unlabeled Diffusion Models, approximating ELBO with labeled and unlabeled data.
result Prevents the generation of sensitive data without compromising image quality.
A framework for sensitivity measures using scoring functions.
problem Constructing sensitivity measures for any elicitable functional.
method Score-based sensitivities constructed via consistent scoring functions.
result Demonstrated intuitive and desirable properties of score-based sensitivities.
This work provides efficient algorithms for approximating ℓ_p sensitivities and related statistics.
problem Estimating the importance of datapoints in high-dimensional datasets.
method Efficient algorithms for computing α-approximation of ℓ_1 sensitivities and total sensitivity using importance sampling and sensitivity computations.
result Real-world datasets have significantly lower intrinsic effective dimensionality than theoretical predictions.