Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

225450675900 · Jun 202019922001200920172026
48 results for sensitivity improvement

Linking output sensitivity to deep learning generalization.

problem Understanding and comparing the generalization properties of deep neural networks.
method Linking the loss function to output sensitivity and analyzing its relation to bias-variance decomposition.
result Output sensitivity is a strong metric for comparing generalization performance of deep networks.

Transformers are less sensitive to input perturbations compared to other models.

problem Understanding the inductive biases of transformers and distinguishing them from other architectures.
method Identified token-wise sensitivity as a metric to explain transformers' inductive biases across different data modalities.
result Transformers have lower sensitivity than MLPs, CNNs, ConvMixers, and LSTMs, across vision and language tasks.

Improved risk-sensitive RL with exponential Bellman equation and better regret bounds.

problem Exponential gap between upper and lower bounds in risk-sensitive RL.
method Identified and addressed deficiencies in existing algorithms and analysis; developed novel analysis and exploration mechanism.
result Improved regret upper bounds over existing ones.

Study risk-sensitive RL in offline settings, improving efficiency and accuracy.

problem Efficiently derive near-optimal policies for risk-sensitive RL using offline data.
method Introduced two provably sample-efficient algorithms for risk-sensitive offline RL in linear MDPs.
result First provably efficient risk-sensitive offline RL algorithms.

Improved bounds for p\ell_p sensitivity sampling reducing the sample complexity for structured matrices.

problem Improving the sample complexity for structured matrices using p\ell_p sensitivity sampling.
method Developed new bounds for p\ell_p sensitivity sampling, achieving a bound of roughly S22/p\mathfrak{S}^{2-2/p} for 2<p<2 < p < \infty.
result Achieved improved bounds for p\ell_p sensitivity sampling, reducing the sample complexity for structured matrices.

Improved subsampling bounds for p\ell_p sensitivity sampling using 2\ell_2 augmentation.

problem Efficiently approximating large data sets by small representative proxies.
method Optimized sampling based on p\ell_p and 2\ell_2 sensitivities.
result Optimal linear ildeO(ε2(S+d)) ilde O(\varepsilon^{-2}(\mathfrak S+d)) sampling complexity for all p[1,2]p \in [1,2].

RISE learns decisions with sensitive variables, improving worst-case outcomes.

problem Uncertainty and bias in decisions due to delayed sensitive variable data.
method Incorporates sensitive variables offline but not at deployment, using quantile or infimum optimization.
result Improves worst-case outcomes for individuals affected by unavailable sensitive variables.

TIER uses extended strain data to improve gravitational wave detection sensitivity.

problem Improving gravitational wave detection sensitivity using extended strain data.
method TIER framework using machine learning to capture extended strain data features.
result Up to 20% improvement in sensitive volume time in LIGO-Virgo-Kagra O3 data.

Improved sensitivity to Higgs potential through neural simulation-based inference for di-Higgs events.

problem Improving sensitivity to physics beyond the Standard Model through di-Higgs events.
method Simulation-based inference using neural networks to estimate per-event likelihood ratios.
result Adding kinematic observables improves experimental sensitivity to Higgs self-coupling.

Improved anti-cancer drug sensitivity prediction using REFINED CNN ensemble learning.

problem Challenges in predicting anti-cancer drug sensitivity for individual cell lines.
method Using REFINED CNN, which represents high-dimensional vectors as compact 2D images with spatial correlations, and building ensembles of these models.
result Ensemble approaches significantly improve drug sensitivity prediction performance compared to single models.

A central goal of algorithmic fairness is to reduce bias in automated decision making. An unavoidable tension exists between accuracy gains obtained by using sensitive information (e.g., gender or ethnic group) as part of a statistical model, and any commitment to protect these characteristics. Often, due to biases pre…

2018-10-19abs ↗pdf ↗

Bayesian approach improves AdaLoRA's performance and efficiency.

problem Improving the efficiency and performance of adaptive low-rank adaptation.
method Utilized Bayesian metrics and the Improved Variational Online Newton (IVON) optimizer for adaptive parameter budget allocation.
result Bayesian counterpart outperforms sensitivity-based importance metric and is faster than AdaLoRA.

New concept of within-group fairness improves AI fairness without sacrificing accuracy.

problem Fairness issues in AI models treating individuals in the same sensitive group unfairly.
method Introducing within-group fairness, proposing mathematical definitions, and developing learning algorithms.
result Improves within-group fairness without sacrificing accuracy and between-group fairness.

The paper introduces gapped scale-sensitive dimensions to improve learning rate bounds.

problem Improving lower bounds on rates of convergence in statistical and online learning.
method Introducing and analyzing gapped scale-sensitive dimensions for function classes.
result Gapped dimensions lead to stronger lower bounds on offset Rademacher averages.

Framework for sensitivity analysis in biomanufacturing processes.

problem High complexity and uncertainty in biomanufacturing processes.
method Shapley value estimation for linear and nonlinear pKG models, using quasi-Monte Carlo and antithetic sampling.
result Improved efficiency and accuracy in sensitivity analysis for biomanufacturing processes.

Global sensitivity analysis improves BNN hyperparameter selection for accurate uncertainty quantification.

problem Difficulties in obtaining accurate uncertainty quantification with Bayesian Neural Networks (BNNs).
method Global sensitivity analysis of BNN performance under varying hyperparameter settings.
result Many hyperparameters interact to affect both predictive accuracy and uncertainty quantification.

Fairness is becoming a rising concern w.r.t. machine learning model performance. Especially for sensitive fields such as criminal justice and loan decision, eliminating the prediction discrimination towards a certain group of population (characterized by sensitive features like race and gender) is important for enhanci…

2019-09-06abs ↗pdf ↗

Risk management in financial derivative markets requires inevitably the calculation of the different price sensitivities. The literature contains an abundant amount of research works that have studied the computation of these important values. Most of these works consider the well-known Black and Scholes model where th…

2017-05-06abs ↗pdf ↗

We consider objective evaluation measures of saliency explanations for complex black-box machine learning models. We propose simple robust variants of two notions that have been considered in recent literature: (in)fidelity, and sensitivity. We analyze optimal explanations with respect to both these measures, and while…

2019-01-27abs ↗pdf ↗

This paper introduces a method to incorporate risk sensitivity in RL using quadratic variation penalties.

problem Risk-sensitive reinforcement learning under entropy regularization.
method Equivalent martingale property and quadratic variation penalty for value process.
result The proposed method improves finite-sample performance in linear-quadratic control problems.

New algorithms improve privacy-preserving data release using external predictions.

problem Privacy-preserving data release with improved utility using external information.
method Learning-augmented algorithms for multiple quantile release.
result Error guarantees scale with prediction quality, almost recovering state-of-the-art guarantees.

New method mitigates bias without sensitive data using causal graph and variational autoencoder.

problem Lack of fairness strategies when sensitive attributes are not collected.
method SRCVAE framework based on causal graph for inferring a proxy sensitive attribute.
result Significant improvements in fairness metrics over existing methods.

Proposes a differentially private bandit algorithm reducing noise over time.

problem Privacy concerns in interactive recommendation systems.
method Tree-based mechanism to add Laplace or Gaussian noise to model parameters, focusing on dynamic global sensitivity.
result Demonstrates (ε,δ)(ε, δ)-differential privacy with reduced noise and improved regret.

A novel method for classification with rejection using ensemble of cost-sensitive classifiers.

problem Avoid risky misclassification in error-critical applications.
method Learning an ensemble of cost-sensitive classifiers.
result Improved classification accuracy and flexibility in loss selection.

Symbolic LSTM improves time series forecasting by reducing hyperparameter sensitivity.

problem High sensitivity to hyperparameters and random initialization in numerical time series forecasting.
method Combining LSTM with a dimension-reducing symbolic representation.
result Symbolic representation alleviates forecasting problems and speeds up training.

BCDP enhances privacy by protecting sensitive features more precisely.

problem Uniform privacy protection in LDP degrades performance for sensitive features.
method Bayesian Coordinate Differential Privacy (BCDP) adjusts privacy protection per feature sensitivity.
result BCDP improves accuracy in downstream tasks without sacrificing privacy.

Proposes a framework to incorporate global sensitivity into local surrogate models.

problem Narrowing focus to local scale in surrogate modeling leads to re-learning global trends.
method Integrates global sensitivity analysis into local surrogate models through input warping.
result Local models become equally sensitive to all input directions, focusing on local dynamics.

Framework improves ETF volatility forecasting by adapting to market conditions.

problem Challenges in volatility forecasting due to shifting market conditions and varying model performance.
method Risk-sensitive specialist routing using online risk-sensitive evaluation and state-dependent gating.
result Reduces forecast loss by 24% and underprediction loss by 22% compared to rolling-best baseline.

Active sampling improves design space exploration for analog circuits.

problem Efficiently exploring the space of design features in analog circuits with many parameters.
method Combining drastic dimension reduction with sensitivity analysis and Bayesian surrogate modeling for active sampling.
result The proposed active sampling flow outperforms traditional Monte-Carlo sampling.