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.

169,291 papers · 148 categories

Trend · papers per month

count % of corpus

8 results for “RRTs”

Rapid Posterior Exploration in Bayesian Non-negative Matrix Factorizationstat.ML

New method uses RRTs to explore Bayesian NMF more effectively.

problem Slow mixing and mode sticking in Bayesian NMF.

method Rapidly exploring random trees (RRTs) in a Bayesian framework.

result Greater coverage of the posterior and higher ELBO values.

Bayesian NMF RRTs Posterior exploration MCMC mixing Non-negative Matrix Factorization

2016-10-27abs ↗pdf ↗

Noise Statistics Oblivious GARD For Robust Regression With Sparse Outliersstat.ML

Noise statistics oblivious algorithm improves robust regression with sparse outliers.

problem Robust regression with sparse outliers in the presence of Gaussian noise.

method Developed a noise statistics oblivious algorithm called RRT-GARD by modifying GARD.

result RRT-GARD performs nearly as well as GARD with known noise statistics.

robust regression sparse outliers noise statistics GARD residual ratio thresholding

2018-09-19abs ↗pdf ↗

Signal and Noise Statistics Oblivious Orthogonal Matching Pursuitstat.ML

New technique RRT improves OMP performance without knowing sparsity or noise.

problem Recovering sparse high-dimensional vectors without knowing sparsity or noise statistics.

method Residual ratio thresholding (RRT) to operate OMP without a priori knowledge.

result RRT achieves comparable performance to OMP with known statistics.

Orthogonal Matching Pursuit Sparse Recovery Residual Ratio Thresholding Noise Statistics Support Recovery

2018-06-02abs ↗pdf ↗

Signal and Noise Statistics Oblivious Sparse Reconstruction using OMP/OLSstat.ML

Develops algorithms for sparse signal reconstruction without needing signal sparsity or noise variance.

problem Sparse signal reconstruction challenges due to unknown signal sparsity and noise variance.

method TF-IGP and RRT-IGP frameworks for OMP and OLS without prior knowledge of

k_0

and

σ^2

result TF-IGP and RRT-IGP achieve successful sparse recovery under restricted isometry conditions.

sparse signal reconstruction OMP/OLS signal sparsity noise variance restricted isometry

2017-07-27abs ↗pdf ↗

Reinforcement Learning with Probabilistically Complete Explorationcs.LG

R3L uses planning algorithms to efficiently explore sparse reward environments.

problem Balancing exploration and exploitation in sparse reward reinforcement learning.

method Formulate exploration as a search problem using RRT, leverage demonstrations from initial solutions to refine RL policy.

result R3L outperforms classic and intrinsic exploration techniques, requiring fewer samples and achieving better asymptotic performance.

reinforcement learning exploration sparse reward planning algorithms RRT

2020-01-20abs ↗pdf ↗

Learning Rational Subgoals from Demonstrations and Instructionscs.AI

Framework learns useful subgoals from demonstrations and instructions.

problem Efficient long-term planning for novel goals.

method Rational subgoals (RSGs) learned from demonstrations and task descriptions.

result Improves performance-time efficiency of planning algorithms.

subgoals planning rational learning demonstrations

2023-03-09abs ↗pdf ↗

Learning to Plan in High Dimensions via Neural Exploration-Exploitation Treescs.LG

NEXT learns efficient paths in high dimensions using neural exploration-exploitation trees.

problem Learning efficient path planning in high-dimensional spaces.

method Neural Exploration-Exploitation Trees (NEXT) integrating neural architecture and UCB algorithm.

result NEXT achieves better sample efficiency and outperforms state-of-the-art methods.

path planning high dimensions neural networks exploration-exploitation sample efficiency

2019-02-28abs ↗pdf ↗

Learning Navigation Costs from Demonstration in Partially Observable Environmentscs.LG

The paper develops a method to learn navigation costs from expert demonstrations in partially observable environments.

problem Learning navigation costs from expert demonstrations in partially observable environments.

method Develops a cost function representation composed of a probabilistic occupancy encoder and a cost encoder, optimized by differentiating the error between demonstrated controls and a control policy computed from the cost encoder.

result The method outperforms baseline IRL algorithms in robot navigation tasks, improving both training and test-time efficiency.

inverse reinforcement learning navigation costs partial observability robot navigation motion planning

2020-02-26abs ↗pdf ↗