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

6 results for “MERLiN”

MERLiN: Mixture Effect Recovery in Linear Networksstat.ME

MERLiN recovers causal signals from observed mixtures in EEG data.

problem Recovering causal signals from observed mixtures in neuroimaging data.

method MERLiN algorithms construct causal variables from non-causal variables using linear combinations.

result MERLiN allows recovery of a cortical signal affected by activity in a brain region, not directly caused by the stimulus.

causal inference linear combinations neuroimaging EEG cortical signals

2015-12-03abs ↗pdf ↗

Merlin: Multi-View Representation Learning for Robust Multivariate Time Series Forecasting with Unfixed Missing Ratescs.LG

Merlin improves robustness of MTSF models to missing data.

problem Suboptimal forecasting performance due to unfixed missing rates in MTSF models.

method Offline knowledge distillation and multi-view contrastive learning.

result Merlin enhances robustness of MTSF models while preserving accuracy.

Time Series Forecasting Deep Learning Missing Data Robustness Representation Learning

2025-06-14abs ↗pdf ↗

Hierarchical Deep Reinforcement Learning Approach for Multi-Objective Scheduling With Varying Queue Sizescs.LG

MERLIN tackles multi-objective task scheduling with hierarchical DRL, outperforming existing methods.

problem Optimizing multiple conflicting constraints in multi-objective task scheduling with varying queue sizes.

method Hierarchical deep reinforcement learning approach to manage large queues efficiently.

result MERLIN outperforms existing methods by a large margin (>22%) on multiple queue sizes.

Hierarchical DRL Multi-objective scheduling Queue management Deep reinforcement learning Optimization

2020-07-17abs ↗pdf ↗

Unsupervised Predictive Memory in a Goal-Directed Agentcs.LG

MERLIN solves tasks with hidden sensor information.

problem Partial observability challenges AI agents.

method MERLIN integrates memory formation with predictive modeling.

result MERLIN solves complex tasks without memory limitations.

MERLIN predictive modeling partial observability memory formation 3D virtual reality

2018-03-28abs ↗pdf ↗

Recovery of non-linear cause-effect relationships from linearly mixed neuroimaging datastat.ME

New algorithm recovers non-linear cause-effect relationships from mixed neuroimaging data.

problem Recovering meaningful cause-effect relationships from linearly mixed neuroimaging data.

method MERLiN (Mixture Effect Recovery in Linear Networks) algorithm, extended to handle non-linear cause-effect relationships.

result The algorithm can recover non-linear cause-effect relationships from linearly mixed neuroimaging data.

causal inference neuroimaging MERLiN non-linear relationships electroencephalography

2016-05-02abs ↗pdf ↗

High-quality nonparallel voice conversion based on cycle-consistent adversarial networkeess.AS

Paper proposes CycleGAN for nonparallel voice conversion.

problem Difficulty in achieving high-quality voice conversion with nonparallel data.

method Cycle-consistent adversarial network (CycleGAN) for nonparallel data-based voice conversion.

result The proposed method outperforms parallel VC methods in subjective evaluations.

voice conversion nonparallel data CycleGAN adversarial network subjective evaluation

2018-04-02abs ↗pdf ↗