arXiv research

Study on protecting sensitive properties of datasets during analysis.

Multi-party machine learning leaks global dataset properties even with black-box access.

Simplified equation predicts model sensitivity to data.

Investigates conditions for risk or utility functionals to be sensitive to large losses.

A framework for sensitivity measures using scoring functions.

Worst-Case Sensitivity measures model sensitivity to uncertainty set size.

We formulate a probabilistic Markov property in discrete time under a dynamic risk framework with minimal assumptions. This is useful for recursive solutions to risk-sensitive versions of dynamic optimisation problems such as optimal prediction, where at each stage the recursion depends on the whole future. The propert…

Gradient-enhanced GSA uses Poincaré chaos expansions for accurate sensitivity analysis.

Linking output sensitivity to deep learning generalization.

Analytical approximations for Asian option sensitivities in Black-Scholes model.

While both cost-sensitive learning and online learning have been studied extensively, the effort in simultaneously dealing with these two issues is limited. Aiming at this challenge task, a novel learning framework is proposed in this paper. The key idea is based on the fusion of online ensemble algorithms and the stat…

Torsion sensitive intersection homology was introduced to unify several versions of Poincare duality for stratified spaces into a single theorem. This unified duality theorem holds with ground coefficients in an arbitrary PID and with no local cohomology conditions on the underlying space. In this paper we consider for…

Cost-Sensitive Online Classification has drawn extensive attention in recent years, where the main approach is to directly online optimize two well-known cost-sensitive metrics: (i) weighted sum of sensitivity and specificity; (ii) weighted misclassification cost. However, previous existing methods only considered firs…

Study examines noise sensitivity of DNNs for binary classification.

Develops algorithms for multi-class Neyman-Pearson classification with cost sensitivity.

New method explains sensitivity of test data uncertainty in Bayesian inference.

Optimal AFs minimize RFR test error and sensitivity.

Efficiently computes optimal policies for Entropic Risk Measures.

tDB removes unfairness from black-box models using probability theory.

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

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

New model shows natural language exhibits phase transition similar to physics.

Image classifiers are sensitive to small changes, affecting most images in a class.

We propose an approach to fair classification that enforces independence between the classifier outputs and sensitive information by minimizing Wasserstein-1 distances. The approach has desirable theoretical properties and is robust to specific choices of the threshold used to obtain class predictions from model output…

Study finds phase transition in context-sensitive language model with short-range interactions.

Deep neural networks (DNNs) have achieved superior performance in various prediction tasks, but can be very vulnerable to adversarial examples or perturbations. Therefore, it is crucial to measure the sensitivity of DNNs to various forms of perturbations in real applications. We introduce a novel perturbation manifold …

Paper proposes a new FRL algorithm for continuous sensitive attributes using EIPM.

Single neuron sensitivity reveals model weaknesses.

In this paper we consider long-run risk sensitive average cost impulse control applied to a continuous-time Feller-Markov process. Using the probabilistic approach, we show how to get a solution to a suitable continuous-time Bellman equation and link it with the impulse control problem. The optimal strategy for the und…

A new jump diffusion regime-switching model is introduced, which allows for linking jumps in asset prices with regime changes. We prove the existence and uniqueness of the solution to the risk-sensitive asset management criterion maximisation problem in this setting. We provide an ODE for the optimal value function, wh…

We propose a cloud-based filter trained to block third parties from uploading privacy-sensitive images of others to online social media. The proposed filter uses Distributed One-Class Learning, which decomposes the cloud-based filter into multiple one-class classifiers. Each one-class classifier captures the properties…

The paper studies risk-sensitive learning schemes and provides learning bounds for empirical OCE minimizers.

Kernel dependence measures yield accurate estimates of nonlinear relations between random variables, and they are also endorsed with solid theoretical properties and convergence rates. Besides, the empirical estimates are easy to compute in closed form just involving linear algebra operations. However, they are hampere…

EXOC framework uses auxiliary variables for counterfactual fairness in machine learning.

Framework for sensitivity analysis in biomanufacturing processes.

Nonlinear similarity measures defined in kernel space, such as correntropy, can extract higher-order statistics of data and offer potentially significant performance improvement over their linear counterparts especially in non-Gaussian signal processing and machine learning. In this work, we propose a new similarity me…

Novel framework for risk-sensitive reinforcement learning with robustness against uncertainty.

This paper adresses the general issue of estimating the sensitivity of the expectation of a random variable with respect to a parameter characterizing its evolution. In finance for example, the sensitivities of the price of a contingent claim are called the Greeks. A new way of estimating the Greeks has been recently i…

New risk class penalizes loss deviations from mean on both sides.

The Gradient Boosting Decision Tree (GBDT) is a popular machine learning model for various tasks in recent years. In this paper, we study how to improve model accuracy of GBDT while preserving the strong guarantee of differential privacy. Sensitivity and privacy budget are two key design aspects for the effectiveness o…

New algorithms reduce risk in reinforcement learning with provable regret bounds.

Many reinforcement learning applications involve the use of data that is sensitive, such as medical records of patients or financial information. However, most current reinforcement learning methods can leak information contained within the (possibly sensitive) data on which they are trained. To address this problem, w…

Researchers develop a new framework to control neural network sensitivity.

Suppose an investor aims at Delta hedging a European contingent claim $h(S(T))$ in a jump-diffusion model, but incorrectly specifies the stock price's volatility and jump sensitivity, so that any hedging strategy is calculated under a misspecified model. When does the erroneously computed strategy super-replicate the t…

New framework enforces demographic parity on distribution tails.

The Wasserstein probability metric has received much attention from the machine learning community. Unlike the Kullback-Leibler divergence, which strictly measures change in probability, the Wasserstein metric reflects the underlying geometry between outcomes. The value of being sensitive to this geometry has been demo…

Develops scalable methods to assess sensitivity and uncertainty in continuous treatment effects.

Study shows explanation disparities in machine learning models are influenced by data and model properties.