C
Charles T. Marx
Researcher at Haverford College
Publications - 9
Citations - 167
Charles T. Marx is an academic researcher from Haverford College. The author has contributed to research in topics: Linear classifier & Multiplicity (mathematics). The author has an hindex of 6, co-authored 9 publications receiving 73 citations.
Papers
More filters
Journal ArticleDOI
Rapid Identification of Marine Plastic Debris via Spectroscopic Techniques and Machine Learning Classifiers.
Anna P. M. Michel,Alexandra E. Morrison,Victoria Preston,Victoria Preston,Charles T. Marx,Beckett C. Colson,Beckett C. Colson,Helen K. White +7 more
TL;DR: Success rates of 76% and higher indicate that ATR-FTIR, NIR reflectance spectroscopy, and LIBS coupled to machine learning classifiers can be used to robustly identify both consumer and environmental plastic samples.
Posted Content
Predictive Multiplicity in Classification
TL;DR: This paper defines predictive multiplicity as the ability of a prediction problem to admit competing models with conflicting predictions, and introduces formal measures to evaluate the severity of predictive multiplier and develops integer programming tools to compute them exactly for linear classification problems.
Proceedings Article
Disentangling Influence: Using disentangled representations to audit model predictions
Charles T. Marx,Richard Lanas Phillips,Sorelle A. Friedler,Carlos Scheidegger,Suresh Venkatasubramanian +4 more
TL;DR: In this paper, the authors propose disentangled influence audits, a procedure to audit the indirect influence of features by identifying proxy features in the dataset, while allowing an explicit computation of feature influence on either individual outcomes or aggregate-level outcomes.
Proceedings Article
Predictive Multiplicity in Classification
TL;DR: In this paper, the authors define predictive multiplicity as the ability of a prediction problem to admit competing models with conflicting predictions and develop integer programming tools to compute them exactly for linear classification problems.
Posted Content
Disentangling Influence: Using Disentangled Representations to Audit Model Predictions
Charles T. Marx,Richard Lanas Phillips,Sorelle A. Friedler,Carlos Scheidegger,Suresh Venkatasubramanian +4 more
TL;DR: It is shown that disentangled representations provide a mechanism to identify proxy features in the dataset, while allowing an explicit computation of feature influence on either individual outcomes or aggregate-level outcomes, and is more powerful than existing methods for ascertaining feature influence.