arXiv research

The paper explores game-theoretic alignment of LLMs with human preferences, finding limitations and conditions.

M3PO improves model-based meta-RL with theoretical guarantees.

The paper analyzes LIME for text data and provides theoretical guarantees.

Machine learning is often used in competitive scenarios: Participants learn and fit static models, and those models compete in a shared platform. The common assumption is that in order to win a competition one has to have the best predictive model, i.e., the model with the smallest out-sample error. Is that necessarily…

We discuss the finite sample theoretical properties of online predictions in non-stationary time series under model misspecification. To analyze the theoretical predictive properties of statistical methods under this setting, we first define the Kullback-Leibler risk, in order to place the problem within a decision the…

The study initiates a theoretical analysis of dynamic benchmarking models.

Theoretical study shows AI models can recover from contaminated training data.

Theoretical analysis of deep neural networks for time series data.

This work explores the generalization properties of diffusion models, providing theoretical and empirical insights.

Bayesian MAML outperforms MAML in meta learning tasks with theoretical guarantees.

The Heston model is validated for option pricing using theoretical derivations and empirical market data.

Consistency models accelerate generation with theoretical guarantees.

The paper establishes theoretical foundations for low-rank knowledge distillation in LLMs.

New model learns from random graph samples to estimate graph parameters.

Theoretical model for iterative user discovery in recommender systems.

New framework evaluates model explanations based on decision task improvement.

We show that model compression can improve the population risk of a pre-trained model, by studying the tradeoff between the decrease in the generalization error and the increase in the empirical risk with model compression. We first prove that model compression reduces an information-theoretic bound on the generalizati…

Proposes a new theoretical framework for PbRL that requires less human feedback.

Theoretical analysis shows MDMs can be efficient but not for all metrics.

New framework for discrete-state diffusion models reduces sample complexity.

The choice of model class is fundamental in statistical learning and system identification, no matter whether the class is derived from physical principles or is a generic black-box. We develop a method to evaluate the specified model class by assessing its capability of reproducing data that is similar to the observed…

This paper analyzes speculative decoding, a method to speed up large language model inferences.

Theoretical study on AI models' resilience to data contamination during recursive training.

This paper establishes a theoretical foundation for consistency training in diffusion models.

Stochastic volatility models describe asset prices $S_t$ as driven by an unobserved process capturing the random dynamics of volatility $σ_t$. Here, we quantify how much information about $σ_t$ can be inferred from asset prices $S_t$ in terms of Shannon's mutual information $I(S_t : σ_t)$. This motivates a careful nume…

This article presents a theoretical model for a dynamic system based on sustainable development. Due to the relatively absence of theoretical studies and practical issues in the area of sustainable development, Romania aspires to the principles of sustainable development. Based on the concept as a process in which econ…

New method uses dynamic programming for meta continual learning.

Model-based reinforcement learning (RL) is considered to be a promising approach to reduce the sample complexity that hinders model-free RL. However, the theoretical understanding of such methods has been rather limited. This paper introduces a novel algorithmic framework for designing and analyzing model-based RL algo…

The paper explores stability and generalization of deep GCNs.

Paper introduces a new generative learning model using Schrödinger bridge diffusion in latent space.

Paper analyzes ECE bias and provides bounds for its estimation.

SyncRank recovers global ranking from noisy comparisons with theoretical guarantees.

Clarifies model-based RL's theoretical issues and counterexamples for popular losses.

Deep neural networks are often used to implement powerful generative models for real-world data. Notable applications include image denoising, as well as other classical inverse problems like compressed sensing and super-resolution. To provide a rigorous but simplified analysis of generative models, in this work, we in…

Information-theoretic Bayesian optimisation techniques have demonstrated state-of-the-art performance in tackling important global optimisation problems. However, current information-theoretic approaches require many approximations in implementation, introduce often-prohibitive computational overhead and limit the choi…

This research provides theoretical guarantees for hyperparameter estimation in complex network dynamical systems.

New method for generating images with conditional probability models.

Boosted additive models reveal new insights and potential pathologies.

Study explores learning behavior of GFlowNets, revealing key mechanisms.

This paper improves generative models by using data scaling and theoretical analysis.

Game-theoretic model captures investor interactions for stock price forecasting.

Unsupervised pre-training improves model generalization, but lacks theoretical understanding.

The paper introduces PD learning to improve deep learning theory.

Diffusion models learn balanced data representations, unlike classification models.

Despite widespread interest and practical use, the theoretical properties of random forests are still not well understood. In this paper we contribute to this understanding in two ways. We present a new theoretically tractable variant of random regression forests and prove that our algorithm is consistent. We also prov…

Although initially originated as a totally empirical relationship to explain the volume of trade between two partners, gravity equation has been the focus of several theoretic models that try to explain it. Specialization models are of great importance in providing a solid theoretic ground for gravity equation in bilat…

Theoretical framework explains why few epochs are enough for LLM fine-tuning.

Theoretical study shows adversarial training improves robustness in deep learning models.