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.

169,341 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “conflict-free”

CYCLADES: Conflict-free Asynchronous Machine Learningstat.ML

CYCLADES speeds up machine learning with conflict-free asynchronous updates.

problem Improving parallel stochastic optimization in shared memory systems.

method Asynchronous parallelization without memory locking, introducing conflict-free updates.

result Consistently outperforms HOGWILD!-type algorithms, achieving up to 5x speedup gains.

machine learning parallel optimization asynchronous conflict-free stochastic algorithms

2016-05-31abs ↗pdf ↗

Resource Allocation for a Wireless Coexistence Management System Based on Reinforcement Learningeess.SP

This paper uses reinforcement learning to predict and allocate resources for coexisting wireless systems.

problem Coexistence of wireless systems in industrial environments leads to resource conflicts.

method Reinforcement learning, specifically deep Q-networks and double deep Q-networks, for resource allocation prediction.

result A prediction accuracy of at least 98% can be achieved in coexistence scenarios.

reinforcement learning resource allocation wireless systems coexistence neural networks

2018-05-24abs ↗pdf ↗