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

4 results for “decontamination”

A Mutual Contamination Analysis of Mixed Membership and Partial Label Modelsstat.ML

Paper addresses decontamination of mixed membership and partial label models.

problem Recovering base distributions from mutual contamination models.

method General setting with arbitrary probability spaces, algorithms for mixed membership and partial label models.

result Necessary and sufficient conditions for identifiability and algorithms for both models.

mutual contamination mixed membership partial label decontamination identifiability

2016-02-19abs ↗pdf ↗

Decontamination of Mutual Contamination Modelsstat.ML

Paper addresses decontamination of mixed distributions in machine learning.

problem Inferring base distributions from mixed samples with label noise.

method General setting with arbitrary probability spaces; sufficient conditions for identifiability; algorithms for infinite and finite samples.

result Sufficient conditions and algorithms for various machine learning problems.

machine learning identifiability label noise multiclass classification demixing

2017-09-30abs ↗pdf ↗

Step and Smooth Decompositions as Topological Clusteringmath.ST

The paper tackles noisy combinations of continuous and step functions, providing conditions for their identification.

problem Recovering noisy observations as a combination of continuous and step functions.

method Topological and local properties of the functions are used to determine conditions for identification. A practical estimation algorithm is provided.

result Conditions for the identification of continuous and step functions based on their global and local properties.

recovery problems step and smooth decompositions topological clustering non-linear ICA graph connectivity

2023-11-09abs ↗pdf ↗

Fairness in Supervised Learning: An Information Theoretic Approachcs.LG

The paper tackles fairness in supervised learning using information theory.

problem Discrimination in decision rules derived from biased historical data.

method Information theoretic framework for designing fair predictors, using equalized odds criterion.

result Designing predictors that are independent of a sensitive attribute while generalizing well.

fairness supervised learning information theory equalized odds discrimination

2018-01-13abs ↗pdf ↗