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.

168,742 papers · 148 categories

Trend · papers per month

count % of corpus

8 results for “descriptor-based”

Why Model Selection Fails in Time Series Forecasting: An Empirical Study of Instability Across Data Regimeseess.SP

Time series forecasting models fail to consistently select the best model across different datasets.

problem Inconsistency in model selection for time series forecasting across varying data regimes.

method Characterized time series using descriptors like trend strength, seasonality, noise level, and temporal dependence. Developed a rule-based selection mechanism to map data regimes to candidate models.

result Rule-based model selection achieves low accuracy, with correct model identification occurring in only a small fraction of cases.

time series forecasting model selection data regimes descriptor-based forecasting performance

2026-05-02abs ↗pdf ↗

Deep Geometric Functional Maps: Robust Feature Learning for Shape Correspondencestat.ML

New method learns shape correspondences robustly from raw geometry.

problem Inaccurate and poor generalization of shape correspondences.

method Learning-based approach with feature-extraction network and functional map representation.

result Robust and accurate shape correspondence learning with less training data.

shape correspondence non-rigid 3D shapes functional maps deep learning geometry-based

2020-03-31abs ↗pdf ↗

Comparison of multi-task convolutional neural network (MT-CNN) and a few other methods for toxicity predictionq-bio.QM

Toxicity analysis and prediction are of paramount importance to human health and environmental protection. Existing computational methods are built from a wide variety of descriptors and regressors, which makes their performance analysis difficult. For example, deep neural network (DNN), a successful approach in many o…

2017-03-31abs ↗pdf ↗

3D Deep Learning for Biological Function Prediction from Physical Fieldsq-bio.BM

Predicting the biological function of molecules, be it proteins or drug-like compounds, from their atomic structure is an important and long-standing problem. Function is dictated by structure, since it is by spatial interactions that molecules interact with each other, both in terms of steric complementarity, as well …

2017-04-13abs ↗pdf ↗

Joint Geometric and Topological Analysis of Hierarchical Datasetscs.LG

Combines geometry and topology for analyzing hierarchical datasets.

problem Analyzing complex, hierarchical datasets with irregular structures.

method Combines manifold learning and topological data analysis.

result Superior classification results compared to state-of-the-art methods.

data analysis topology geometry hierarchical datasets manifold learning

2021-04-03abs ↗pdf ↗

Tiled Squeeze-and-Excite: Channel Attention With Local Spatial Contextstat.ML

Tiled Squeeze-and-Excite improves channel attention with local spatial context.

problem Improving channel attention mechanisms in neural networks.

method Proposes tiled squeeze-and-excite (TSE) framework for channel attention.

result Local context of 7 rows or columns is sufficient for matching global context performance.

channel attention squeeze-and-excite local context global context AI accelerators

2021-07-05abs ↗pdf ↗

Fast Estimation of Information Theoretic Learning Descriptors using Explicit Inner Product Spacescs.LG

Improved ITL descriptors using explicit inner product spaces for scalable systems.

problem Scalability issues in ITL due to high computational complexity.

method Explicit inner product space (EIPS) kernels for ITL, leveraging data-independent basis.

result Superior performance of EIPS-ITL estimators and combined NT-KAF using EIPS-ITL cost functions.

ITL Kernel methods Scalability Renyi entropy Reproducing kernel Hilbert space

2020-01-01abs ↗pdf ↗

Benchmarking unsupervised near-duplicate image detectioncs.CV

Unsupervised near-duplicate detection has many practical applications ranging from social media analysis and web-scale retrieval, to digital image forensics. It entails running a threshold-limited query on a set of descriptors extracted from the images, with the goal of identifying all possible near-duplicates, while l…

2019-07-03abs ↗pdf ↗