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

22446587 · May 202619922001200920172026
48 results for KL inequality

Paper relaxes triangle inequality for KL divergence between Gaussian distributions.

problem KL divergence does not satisfy triangle inequality for Gaussian distributions.
method Investigates relaxed triangle inequality and finds supremum.
result Supremum of KL divergence is found and conditions for attaining it are determined.

The paper explores how information geometry impacts classical CR inequalities.

problem Deriving and generalizing CR inequalities using information geometry.
method Examining Eguchi's theory and applying Amari-Nagoaka's theory to KL-divergence, and then extending to other divergences.
result Generalized CR inequalities derived from various divergences.

New schemes improve error estimates for sampling from non-log-concave distributions.

problem Improving sampling from non-log-concave distributions with super-linear drift growth.
method Developed tamed Euler and randomized Euler schemes with error estimates.
result Near-optimal error bounds for sampling and optimization problems.

New bounds close the score matching gap for diffusion models.

problem The difference between sample quality and score matching loss in diffusion models.
method Theoretical analysis of score matching gap, developing tighter bounds for KL divergence, reverse KL divergence, and Wasserstein distance.
result The quality of score approximation impacts closing the score matching gap for low noise scales.

New method improves sampling for weakly log-concave posteriors.

problem Sampling from weakly log-concave posterior distributions.
method Stochastic Langevin Monte Carlo with over-damped diffusion.
result Simulation horizon is (dlog(n)2)(1+r)2(d \log(n)^2)^{(1+r)^2} with Poisson subsampling.

Method identifies low-dimensional structure in high-dimensional probability measures.

problem Identifying low-dimensional structure in high-dimensional probability measures.
method Extends prior work on minimizing majorizations of the Kullback-Leibler divergence to identify optimal approximations within a specific class of measures.
result Connection between dimensional logarithmic Sobolev inequality and approximations with the ansatz.

Unified analysis of KL divergence using shifted composition for sampling.

problem Sampling from target distributions with KL divergence guarantees.
method Shifted composition rule applied to KL divergence, combining local error analysis and Girsanov's theorem.
result Unified KL guarantees for strongly log-concave, weakly log-concave, and log-Sobolev distributions.

The study explores geodesics and KL-divergence on Hölder equilibrium probabilities.

problem Finding the probability that minimizes KL-divergence from a fixed probability in a convex set of probabilities.
method Analyzes geodesics paths on the manifold of Hölder equilibrium probabilities and uses KL-divergence as a metric.
result Explicit equations for the solution of the minimization problem are derived.

Paper improves convergence rate of Langevin Dynamics algorithms.

problem Sampling problems and non-convex optimization in machine learning.
method Stochastic Variance Reduced Gradient Langevin Dynamics and Stochastic Recursive Gradient Langevin Dynamics with improved convergence rates.
result Proves convergence to objective distribution under weaker conditions.

We study the Proximal Langevin Algorithm (PLA) for sampling from a probability distribution ν=efν= e^{-f} on Rn\mathbb{R}^n under isoperimetry. We prove a convergence guarantee for PLA in Kullback-Leibler (KL) divergence when νν satisfies log-Sobolev inequality (LSI) and ff has bounded second and third derivatives. Thi…

2019-11-04abs ↗pdf ↗

The paper analyzes the reward improvement of aligned policies in large language models.

problem Optimizing policies in large language models while staying close to a reference policy.
method Information-theoretic analysis and reduction to exponential order statistics.
result Information-theoretic upper bounds on reward improvement are derived.

Universal tester-learner for halfspaces over structured distributions.

problem Learning halfspaces over a wide class of structured distributions.
method Uses a fully polynomial tester-learner based on hypercontractivity and sum-of-squares (SOS) programs.
result Achieves error O(opt)+εO(\mathrm{opt}) + ε on any labeled distribution that the tester accepts.

Paper resolves bias in ALFT training using generalized alignment games.

problem Systematic bias in estimating logarithmic rewards from small batches.
method Generalized Distributional Alignment Games, U-statistics, minimax polynomial estimators, Variance-Optimal Augmented Polynomial Optimization Program (AQP) Estimator.
result Proves optimal bias and accelerated convergence in ALFT training.

Tail-Safe hedging uses reinforcement learning with a safety layer to manage financial risks.

problem Managing financial risks in derivatives trading with robustness and explainability.
method Combines distributional reinforcement learning with a CBF-QP safety layer to enforce financial constraints.
result Improves risk management without degrading central performance and avoids hard constraint violations.

Paper proposes an algorithm for sampling from complex mixture distributions without requiring smoothness.

problem Sampling from a mixture of weakly smooth potentials.
method Unadjusted Langevin algorithm with Euler discretization for a mixture of weakly smooth distributions.
result Convergence in Kullback-Leibler divergence and LβL_β-Wasserstein metric with polynomial dependence on dimension.

Improved Langevin algorithms with prior diffusion achieve dimension-independent convergence for non-log-concave distributions.

problem Understanding the dimension dependency of computational complexity in high-dimensional sampling.
method Investigation of prior diffusion technique for log-Sobolev inequality target distributions.
result Modified Langevin algorithm achieves dimension-independent KL divergence convergence.

TSC uses HMC and adaptive transport maps to optimize forward KL for variational inference.

problem Variational inference underestimates uncertainty when minimizing reverse KL.
method TSC uses Hamiltonian Monte Carlo and adaptive transport maps to optimize KL(p||q).
result TSC achieves competitive performance in training variational autoencoders on large-scale data.

M-FISHER detects and adapts to streaming data shifts with statistical validity and stability.

problem Detecting and adapting to distributional shifts in streaming data.
method Constructs an exponential martingale from non-conformity scores and applies Ville's inequality for detection. Fisher-preconditioned updates for adaptation.
result Establishes M-FISHER as a principled approach for robust, anytime-valid detection and geometrically stable adaptation.

A classic setting of the stochastic K-armed bandit problem is considered in this note. In this problem it has been known that KL-UCB policy achieves the asymptotically optimal regret bound and KL-UCB+ policy empirically performs better than the KL-UCB policy although the regret bound for the original form of the KL-UCB…

2019-03-19abs ↗pdf ↗

A new variational inference method using sliced Wasserstein distance is proposed.

problem The inefficiency and unreasonable properties of Kullback-Leibler divergence.
method Minimizing sliced Wasserstein distance, a valid metric from optimal transport.
result The proposed method approximates the unnormalized distribution efficiently and without requiring a tractable density function.

Improved fast rates for decision making with forward-KL regularization in contextual bandits.

problem Improving fast rates for decision making with forward-KL regularization in contextual bandits.
method Streamlined analysis of forward-KL-regularized offline CBs, exploiting the pessimism principle and convex-analytical pipeline.
result First ildeO(ε1) ilde{O}(ε^{-1}) upper bounds in tabular and general function approximation settings.

We present a new PAC-Bayesian generalization bound. Standard bounds contain a $\sqrt{L_n \cdot \KL/n}$ complexity term which dominates unless LnL_n, the empirical error of the learning algorithm's randomized predictions, vanishes. We manage to replace LnL_n by a term which vanishes in many more situations, essentially …

2019-05-31abs ↗pdf ↗

Kurdyka-Lojasiewicz (KL) exponent plays an important role in estimating the convergence rate of many contemporary first-order methods. In particular, a KL exponent of 12\frac12 for a suitable potential function is related to local linear convergence. Nevertheless, KL exponent is in general extremely hard to estimate. I…

2019-02-10abs ↗pdf ↗

The paper develops new algorithms for KL-divergence NMF, proving convergence and performance.

problem Improving NMF for nonnegative data with KL divergence.
method Collect and analyze properties of KL objective function, propose and test new algorithms.
result Guaranteed non-increasing objective function for one proposed algorithm, global convergence.

This paper tightens the law of the iterated logarithm for empirical KL_inf, applicable to unbounded data.

problem Developing nonasymptotic concentration bounds for empirical KL_inf with optimal constants and rates.
method Presenting a tight law of the iterated logarithm for empirical KL_inf, applicable to unbounded data.
result A tight law of the iterated logarithm for empirical KL_inf, applicable to unbounded data.

Paper analyzes risk bounds for in-context learning in multiclass classification.

problem Risk bounds for in-context learning in multiclass classification.
method Formalizes tasks as sequences of labeled examples and queries, estimates conditional class probabilities, establishes oracle inequality for KL divergence.
result ICL achieves minimax optimal rate for conditional probability estimation.

Paper analyzes and improves KL-regularized RL for LLMs with logarithmic regret.

problem Improving efficiency of RL fine-tuning for large language models.
method Optimism-based KL-regularized online contextual bandit algorithm with novel regret analysis.
result Achieves an O(ηlog(NRT)dR)\mathcal{O}\big(η\log (N_{\mathcal R} T)\cdot d_{\mathcal R}\big) logarithmic regret bound.

Theory for RLHF generalization under reward shift and clipped KL.

problem Theoretical understanding of RLHF generalization, especially with reward shift and clipped KL.
method Developed generalization theory for RLHF, accounting for reward shift and clipped KL.
result Presented generalization bounds for RLHF, suggesting generalization error from sampling, reward shift, and KL clipping.