arXiv research

The paper addresses biased preferences in candidate selection, proposing a fair and utility-maximizing algorithm.

The Mutual Fund Theorem (MFT) is considered in a general semimartingale financial market S with a finite time horizon T, where agents maximize expected utility of terminal wealth. It is established that: 1) Let N be the wealth process of the numéraire portfolio (i.e. the optimal portfolio for the log utility). If any p…

New method for fair resource allocation in AI-aware networks with unknown utility functions.

Study optimizes fairness in predictive models by balancing utility and separation.

Complex classification performance metrics such as the F${}_β$-measure and Jaccard index are often used, in order to handle class-imbalanced cases such as information retrieval and image segmentation. These performance metrics are not decomposable, that is, they cannot be expressed in a per-example manner, which hinder…

We study the convex duality method for robust utility maximization in the presence of a random endowment. When the underlying price process is a locally bounded semimartingale, we show that the fundamental duality relation holds true for a wide class of utility functions on the whole real line and unbounded random endo…

A new model shows fairness mechanisms can improve selection utility even without implicit bias.

In the general framework of a semimartingale financial model and a utility function $U$ defined on the positive real line, we compute the first-order expansion of marginal utility-based prices with respect to a ``small'' number of random endowments. We show that this linear approximation has some important qualitative …

Paper tackles non-monotonic resource utilization in sequential decision-making.

We consider the robust exponential utility maximization problem in discrete time: An investor maximizes the worst case expected exponential utility with respect to a family of nondominated probabilistic models of her endowment by dynamically investing in a financial market, and statically in available options. We show …

Paper proposes new strategies for better portfolio estimation in long-term investments with unknown distributions.

In this paper we ask whether, given a stock market and an illiquid derivative, there exists arbitrage-free prices at which an utility-maximizing agent would always want to buy the derivative, irrespectively of his own initial endowment of derivatives and cash. We prove that this is false for any given investor if one c…

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

To any utility maximization problem under transaction costs one can assign a frictionless model with a price process $S^*$, lying in the bid/ask price interval $[\underline S, \bar{S}]$. Such process $S^*$ is called a \emph{shadow price} if it provides the same optimal utility value as in the original model with bid-as…

Study learns linear utility functions from comparisons, showing learnability gaps between passive and active learning.

Paper uses ensemblers to predict sepsis early from patient records.

We study the two-times differentiability of the value functions of the primal and dual optimization problems that appear in the setting of expected utility maximization in incomplete markets. We also study the differentiability of the solutions to these problems with respect to their initial values. We show that the ke…

Specifying utility functions is a key step towards applying the discrete choice framework for understanding the behaviour processes that govern user choices. However, identifying the utility function specifications that best model and explain the observed choices can be a very challenging and time-consuming task. This …

Derives an approximate solution for power utility optimization under predictable returns.

Current approaches in approximate inference for Bayesian neural networks minimise the Kullback-Leibler divergence to approximate the true posterior over the weights. However, this approximation is without knowledge of the final application, and therefore cannot guarantee optimal predictions for a given task. To make mo…

Study privacy-utility trade-off in IoT time-series data sharing with RL.

The paper resolves a counterexample showing convergence of expected utility in binomial models.

We study the robustness of active learning (AL) algorithms against prior misspecification: whether an algorithm achieves similar performance using a perturbed prior as compared to using the true prior. In both the average and worst cases of the maximum coverage setting, we prove that all $α$-approximate algorithms are …

New framework for learning from various weak supervision types.

Study improves stock return uncertainty prediction using Gaussian mixture distributions.

Better data representations can simplify learning tasks by aligning model distributions with true data distributions.

New framework uses background knowledge to speed up causal discovery.

User releases data to service provider while balancing privacy and utility.

A new learning method uses data to learn from large model sets.

Two methods estimate effect size for online experiments, improving accuracy and efficiency.

In cases of uncertainty, a multi-class classifier preferably returns a set of candidate classes instead of predicting a single class label with little guarantee. More precisely, the classifier should strive for an optimal balance between the correctness (the true class is among the candidates) and the precision (the ca…

The unprecedented demand for large amount of data has catalyzed the trend of combining human insights with machine learning techniques, which facilitate the use of crowdsourcing to enlist label information both effectively and efficiently. The classic work on crowdsourcing mainly focuses on the label inference problem …

A new method calculates optimal decisions from classifier outputs, improving predictions in drug discovery.

Paper introduces Isotonic Mechanism for better item scoring.

This work addresses privacy issues in IoT data sharing by balancing information disclosure and user privacy.

Bayesian nonparametric LABS model adapts to function smoothness in Besov spaces.

SALSA efficiently approximates leverage scores for big data, improving ARMA model fitting.

Optimizes long-term social welfare in recommender systems by matching users to providers.

In typical applications of Bayesian optimization, minimal assumptions are made about the objective function being optimized. This is true even when researchers have prior information about the shape of the function with respect to one or more argument. We make the case that shape constraints are often appropriate in at…

We propose a novel sparse preference learning/ranking algorithm. Our algorithm approximates the true utility function by a weighted sum of basis functions using the squared loss on pairs of data points, and is a generalization of the kernel matching pursuit method. It can operate both in a supervised and a semi-supervi…

New synthetic data analysis reveals high type 1 error rates.

Nonparametric adaptive robust control tackles model uncertainty in stochastic processes.

PH-CS selects test inputs with reliability guarantees, adapting FDR to data.

Generative model improves safety in self-driving simulators and human motion generation.

In this paper we introduce kinetic equations for the evolution of the probability distribution of two goods among a huge population of agents. The leading idea is to describe the trading of these goods by means of some fundamental rules in price theory, in particular by using Cobb-Douglas utility functions for the bina…

Consider an agent taking two successive decisions to maximize his expected utility under uncertainty. After his first decision, a signal is revealed that provides information about the state of nature. The observation of the signal allows the decision-maker to revise his prior and the second decision is taken according…

Link prediction is one of the fundamental problems in network analysis. In many applications, notably in genetics, a partially observed network may not contain any negative examples of absent edges, which creates a difficulty for many existing supervised learning approaches. We develop a new method which treats the obs…

Optimizes portfolio in volatile markets with jumps, providing accurate formulas.