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.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

5 results for “optimization-driven”

L2M: Practical posterior Laplace approximation with optimization-driven second moment estimationcs.LG

A new method for uncertainty estimation in neural networks using existing optimization steps.

problem Uncertainty quantification in deep neural networks.

method L2M: Practical posterior Laplace approximation with optimization-driven second moment estimation.

result L2M method yields reasonable results without requiring changes in models or extra computational steps.

uncertainty quantification deep neural networks Laplace approximation second moment optimization-driven

2021-07-09abs ↗pdf ↗

Target search optimization by threshold resettingcond-mat.stat-mech

Optimizes search times by resetting agents when a threshold is reached.

problem Improving search efficiency in systems with thresholds.

method Develops a framework for correlated stochastic processes with threshold resetting.

result Optimal resetting can prevent larger losses and is applicable to various stochastic systems.

threshold resetting first passage time stochastic processes search optimization correlated resets

2025-04-18abs ↗pdf ↗

Minimum Volume Conformal Sets for Multivariate Regressionstat.ML

Optimizes minimum-volume prediction sets for multivariate regression.

problem Lack of efficient methods for multivariate conformal prediction.

method Optimization-driven framework for minimum-volume covering sets.

result Efficient and informative prediction sets with tight coverage.

conformal prediction multivariate regression minimum volume optimization valid coverage

2025-03-24abs ↗pdf ↗

FLAG-Trader: Fusion LLM-Agent with Gradient-based Reinforcement Learning for Financial Tradingcs.AI

Unified LLM-agent with RL improves financial trading performance.

problem LLMs struggle with complex, multi-step financial tasks.

method Fusion of LLMs and gradient-based RL for policy optimization.

result Improves LLM performance in trading and other financial tasks.

LLM Reinforcement Learning Financial Trading Policy Optimization Gradient Descent

2025-02-17abs ↗pdf ↗

FLoE: Fisher-Based Layer Selection for Efficient Sparse Adaptation of Low-Rank Expertscs.LG

FLoE adapts LLMs by selectively deploying LoRA adapters based on layer importance and task requirements.

problem Uniform LoRA deployment across all layers leads to inefficient and redundant parameter allocation.

method FLoE uses Fisher information to dynamically identify task-critical layers and optimizes LoRA ranks.

result FLoE achieves significant efficiency-accuracy trade-offs, especially in resource-constrained environments.

PEFT LoRA Fisher information MoE Bayesian optimization

2025-05-31abs ↗pdf ↗