arXiv research

Efficient algorithm for global optimization of multivariate Lipschitz functions.

New algorithm optimizes Hölder continuous functions efficiently.

Study efficient sequential evaluation of large language models using historical data.

[Context:] Model-based testing is an instrument for automated generation of test cases. It requires identifying requirements in documents, understanding them syntactically and semantically, and then translating them into a test model. One light-weight language for these test models are Cause-Effect-Graphs (CEG) that ca…

A new stochastic method handles ensemble creation with cost constraints.

Recently, in order to explore the mechanism behind wealth or income distribution, several models have been proposed by applying principles of statistical mechanics. These models share some characteristics, such as consisting of a group of individual agents, a pile of money and a specific trading rule. Whatever the trad…

We propose scalable methods to execute counting queries in machine learning applications. To achieve memory and computational efficiency, we abstract counting queries and their context such that the counts can be aggregated as a stream. We demonstrate performance and scalability of the resulting approach on random quer…

New bounds show complex neural networks need many queries to learn.

sGPO optimizes RLVR training by balancing inference FLOPs and training efficiency.

New insights into query complexity for Nash equilibrium learning.

Platform uses queries to elicit investor preferences for portfolio trades, improving allocation efficiency.

We consider the following conditional linear regression problem: the task is to identify both (i) a $k$-DNF condition $c$ and (ii) a linear rule $f$ such that the probability of $c$ is (approximately) at least some given bound $μ$, and $f$ minimizes the $\ell_p$ loss of predicting the target $z$ in the distribution of …

Active inference framework improves $U$-statistic estimation efficiency.

The growth of the modern knowledge-based economy is becoming less and less dependent on tangible assets and more on intangible ones. In this context, the role of human capital in the value creation process has become central. Despite the large amount of scientific work on human capital phenomena, little research has re…

Assisted by neural networks, reinforcement learning agents have been able to solve increasingly complex tasks over the last years. The simulation environment in which the agents interact is an essential component in any reinforcement learning problem. The environment simulates the dynamics of the agents' world and henc…

This review is about the convenience, the benefits, as well as the destructive capacities of money. It deals with various aspects of money creation, with its value, and its appropriation. All sorts of money tend to get corrupted by eventually creating too much of them. In the long run, this renders money worthless and …

Unified framework controls false discovery rate in bandit multiple testing.

Study compares methods for improving document retrieval accuracy.

The paper optimizes k-NN for distributed learning with minimax optimal performance.

ReSkill reconciles RL skill creation with policy optimization.

New KD-tree based method for private synthetic data generation.

Survey on AL strategies for cost-effective annotation in classification.

The distribution of firms' growth and firms' sizes is a topic under intense scrutiny. In this paper we show that a thermodynamic model based on the Maximum Entropy Principle, with dynamical prior information, can be constructed that adequately describes the dynamics and distribution of firms' growth. Our theoretical fr…

New optimal transport method handles mass creation and destruction.

The study sets limits on how well halfspaces can be learned when labels are corrupted.

Designing AI market for content creation

Avaya Conversational Intelligence(ACI) is an end-to-end, cloud-based solution for real-time Spoken Language Understanding for call centers. It combines large vocabulary, real-time speech recognition, transcript refinement, and entity and intent recognition in order to convert live audio into a rich, actionable stream o…

Proposes a model to optimize feedback for content creators on social media.

Pure exploration (aka active testing) is the fundamental task of sequentially gathering information to answer a query about a stochastic environment. Good algorithms make few mistakes and take few samples. Lower bounds (for multi-armed bandit models with arms in an exponential family) reveal that the sample complexity …

Unified framework for deferring queries to top-k experts, improving accuracy-cost trade-offs.

LLMs translate natural language trading intents into correct option strategies using a domain-specific language.

New lower bounds show learning intersections of halfspaces is hard even for a few halfspaces.

Proposes RPG-RT for red-teaming T2I models without internal access.

This research creates and classifies datasets for Setswana and Sepedi news headlines.

This paper addresses detection of a reverse engineering (RE) attack targeting a deep neural network (DNN) image classifier; by querying, RE's aim is to discover the classifier's decision rule. RE can enable test-time evasion attacks, which require knowledge of the classifier. Recently, we proposed a quite effective app…

This paper gives an overview of several key innovations in the 19th century which led to complex geometry in the 20th century. This includes the creation of the complex plane, the work of Abel on addition theorems for generalized elliptic integrals, the theory of elliptic functions, holomorphic functions, and the creat…

We study the possibility of completing data bases of a sample of governance, diversification and value creation variables by providing a well adapted method to reconstruct the missing parts in order to obtain a complete sample to be applied for testing the ownership-structure/diversification relationship. It consists o…

In the learning to learn (L2L) framework, we cast the design of optimization algorithms as a machine learning problem and use deep neural networks to learn the update rules. In this paper, we extend the L2L framework to zeroth-order (ZO) optimization setting, where no explicit gradient information is available. Our lea…

The extremely useful method of Malliavin calculus has not yet gained adequate popularity because of the complicated analytic apparatus of this method. The author attempts here to propose a simplified algebraic formalism similar to Malliavin calculus, but based on the notion of creation-annihilation operators instead of…

Model explains money creation under regulatory constraints.

The study examines how many samples are needed to minimize a noisy convex function with inaccurate gradient estimates.

Optimal experiments tighten causal effect bounds efficiently.

Bayesian optimization is a sample-efficient approach to global optimization that relies on theoretically motivated value heuristics (acquisition functions) to guide its search process. Fully maximizing acquisition functions produces the Bayes' decision rule, but this ideal is difficult to achieve since these functions …

It has been widely understood that differential privacy (DP) can guarantee rigorous privacy against adversaries with arbitrary prior knowledge. However, recent studies demonstrate that this may not be true for correlated data, and indicate that three factors could influence privacy leakage: the data correlation pattern…

Solve Painleve VI to relate instanton bundles.

Solves TOD systems' query annotation problem without explicit annotations.

The paper tackles sequential mode estimation with oracle queries.

Study exact community recovery in noisy SBM with limited queries.