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.

169,051 papers · 148 categories

Trend · papers per month

79157236314 · Jun 202019922001200920182026
48 results for Direct Detection

Improves community detection in directed networks with theoretical guarantees.

problem Degree heterogeneity affects community detection in directed networks.
method Introduced D-SCORE algorithm and established theoretical guarantees for Directed-DCBM.
result Established theoretical guarantees and provided improvements for D-SCORE.

The paper presents a method for sound event localization and detection using CRNN models.

problem Sound event localization and detection in complex environments.
method Consecutive ensemble of CRNN models for estimating event onset, offset, direction of arrival, and classification.
result The proposed method outperforms other participants in the DCASE2019 task3.

Bi-directional Curriculum Learning improves graph anomaly detection by considering both homogeneity and heterogeneity.

problem Existing graph anomaly detection methods often ignore the different contributions of nodes to training.
method Introduces Bi-directional Curriculum Learning (BCL) to optimize GAD methods by considering both homogeneity and heterogeneity of nodes.
result Extensive experiments show that BCL significantly improves the performance of GAD anomaly detection models.

ZDP detects drift in large language models without labels, proving key theorems and metrics.

problem Detecting drift in large language models without task labels or output evaluations.
method Zero-Direction Probing (ZDP) framework based on null directions of transformer activations, proving theoretical guarantees.
result Proves the Variance--Leak Theorem, Fisher Null-Conservation, Rank--Leak bound, and logarithmic-regret guarantee.

Detects adversarial directions to make reinforcement learning policies more robust.

problem Adversarial attacks exploit non-robust directions in reinforcement learning policies, leading to instability.
method Local quadratic approximation of deep neural policy loss to identify non-robust directions.
result Provides a theoretical basis for distinguishing safe from adversarial observations.

A new framework uses information theory to detect anomalies in images without labeled data.

problem Detect anomalies in images without labeled data.
method A direct objective function using information theory to maximize the distance between normal and anomalous data.
result The proposed framework significantly outperforms state-of-the-arts on multiple benchmark datasets.

Unsupervised method discovers interpretable directions in GAN latent space.

problem Discovering interpretable directions in GAN latent space without supervision.
method Model-agnostic procedure to identify directions corresponding to semantic manipulations.
result Findings include directions for background removal and competitive saliency detection performance.

Machine learning detects drilling anomalies, reducing accidents and costs.

problem Detecting and preventing accidents during directional drilling.
method Time-series comparison using machine learning and Gradient Boosting classification.
result The model detects half of the anomalies with about 0.53 false alarms per day.

Generalized change-point detection using various binary models.

problem Discovering changes in time series distribution.
method Direct density ratio estimation with Gradient Boosting over Decision Trees and Neural Networks.
result Proposed methods outperform classical RuLSIF algorithm.

New method for community detection in sparse directed SBMs with exact recovery guarantees.

problem Exact recovery in sparse directed SBMs, especially with growing communities.
method Two-stage procedure: neighborhood-smoothing followed by KK-means clustering.
result Exact recovery of all community labels with probability tending to one under mild sparsity and separation conditions.

Online CPD for weighted and directed graphs using RDPG model.

problem Monitoring and detecting changes in weighted and directed graph data.
method Spectral embeddings of RDPG models for online updates and error-rate control.
result A lightweight online CPD algorithm with improved detection resolution and delay.

We show that the cobordism groups of negative codimensional folds maps contain direct sums of stable homotopy groups of Thom spaces of vector bundles like the circle and the infinite dimensional projective space. We give geometrical invariants which detect these direct summands.

2007-04-24abs ↗pdf ↗

Survey of deep learning applications in traffic congestion detection, prediction, and alleviation.

problem Improving transportation network service by detecting, predicting, and alleviating traffic congestion.
method Comprehensive review of deep learning applications in transportation, focusing on challenges specific to congestion.
result Identification of challenges and gaps in current research on deep learning for traffic congestion.

Study assesses risk of upward lightning at wind turbines using direct measurements and machine learning.

problem Risk underestimation of upward lightning at wind turbines due to limited detection by current standards.
method Direct UL measurements linked to meteorological reanalysis data using random forests.
result Risk maps based on case study events show high probabilities coincide with actual UL events.

Proposes TNPM for better node popularity in directed and bipartite networks.

problem Lack of node popularity consideration in community detection of directed and bipartite networks.
method Two-Way Node Popularity Model (TNPM) with Delete-One-Method (DOM) and Two-Stage Divided Cosine Algorithm (TSDC).
result Improved estimation accuracy and computational efficiency demonstrated through real-world applications.

Novel technique detects adversarial samples in face recognition models.

problem Widespread adversarial sample attacks on DNN models.
method Bi-directional correspondence inference between attributes and internal neurons to identify critical neurons.
result 94% detection accuracy for 7 different kinds of attacks with 9.91% false positives.

A drone-based MOT algorithm tracks vehicles using neural network detections and TPMBM filter.

problem Tracking multiple vehicles from drone-mounted cameras.
method Neural network for object detection, TPMBM filter for trajectory estimation, von-Mises Fisher distribution for DOA.
result TPMBM filter optimally estimates vehicle trajectories.

Novel CGTF model for recommender systems and community detection from coupled graphs and tensors.

problem Lack of effective methods for analyzing multiple information repositories with graph side information.
method Coupled Graph-Tensor Factorization (CGTF) with ADMM for nonnegative factor recovery.
result CGTF model successfully detects communities even with missing graph links.

Detects underspecification in pre-trained models using local ensembles.

problem Underspecification in pre-trained models where many predictors are consistent with training data.
method Uses local second-order information to approximate prediction variance across an ensemble of models.
result Capable of detecting underspecification in pre-trained models on test data.

This is a survey paper of author's results on cobordism groups and semigroups of fold maps and simple fold maps. The results include: establishing a relation between fold maps and immersions through geometrical invariants of cobordism classes of fold maps and simple fold maps in terms of immersions with prescribed norm…

2007-09-04abs ↗pdf ↗

We introduce a new unsupervised anomaly detection ensemble called SPI which can harness privileged information - data available only for training examples but not for (future) test examples. Our ideas build on the Learning Using Privileged Information (LUPI) paradigm pioneered by Vapnik et al. [19,17], which we extend …

2018-05-06abs ↗pdf ↗

Detects malicious accounts in permissionless blockchains using graph properties and ML.

problem Identifying and classifying malicious accounts in permissionless blockchains.
method Temporal graph properties, ML algorithms (ExtraTreesClassifier, K-Means), cosine similarity, behavior change analysis.
result ExtraTreesClassifier performs best in detecting malicious accounts on Ethereum blockchain.

We accelerate Bayesian inference for neutrino physics experiments by 100-60x.

problem Complex posterior geometries in multi-dimensional parameter spaces.
method GPU acceleration, automatic differentiation, neural-network-guided reparameterization.
result Significant performance improvements in Bayesian inference for direct detection experiments.

Constructs Lie algebras from labeled directed graphs and identifies properties of these algebras.

problem Constructing and analyzing Lie algebras from labeled directed graphs.
method Using labeled directed simple graphs to construct 2-step nilpotent Lie algebras, identifying ideals and subalgebras through special subgraphs, and proving isomorphisms based on label occurrences.
result Lie algebras depend only on the underlying undirected graph if all edges are labeled uniquely.