Get our free extension to see links to code for papers anywhere online!

Add to Chrome

Add to Firefox

Picture for Eric Heim

Eric Heim

University of Pittsburgh

What is Your Metric Telling You? Evaluating Classifier Calibration under Context-Specific Definitions of Reliability

May 23, 2022
John Kirchenbauer, Jacob Oaks, Eric Heim

Figure 1 for What is Your Metric Telling You? Evaluating Classifier Calibration under Context-Specific Definitions of Reliability

Figure 2 for What is Your Metric Telling You? Evaluating Classifier Calibration under Context-Specific Definitions of Reliability

Figure 3 for What is Your Metric Telling You? Evaluating Classifier Calibration under Context-Specific Definitions of Reliability

Figure 4 for What is Your Metric Telling You? Evaluating Classifier Calibration under Context-Specific Definitions of Reliability

* Accepted in the ICLR 2022 Machine Learning Evaluation Standards Workshop

Access Paper or Ask Questions

Measuring AI Systems Beyond Accuracy

Apr 07, 2022
Violet Turri, Rachel Dzombak, Eric Heim, Nathan VanHoudnos, Jay Palat, Anusha Sinha

* 8 pages, Presented at 2022 AAAI Spring Symposium Series Workshop on AI Engineering: Creating Scalable, Human-Centered and Robust AI Systems

Access Paper or Ask Questions

Factor Analysis on Citation, Using a Combined Latent and Logistic Regression Model

Dec 02, 2019
Namjoon Suh, Xiaoming Huo, Eric Heim, Lee Seversky

Figure 1 for Factor Analysis on Citation, Using a Combined Latent and Logistic Regression Model

Figure 2 for Factor Analysis on Citation, Using a Combined Latent and Logistic Regression Model

Figure 3 for Factor Analysis on Citation, Using a Combined Latent and Logistic Regression Model

Figure 4 for Factor Analysis on Citation, Using a Combined Latent and Logistic Regression Model

* Citation network, matrix decomposition, latent variable model, logistic regression model, convex optimization, alternating direction method of multiplier

Access Paper or Ask Questions

xBD: A Dataset for Assessing Building Damage from Satellite Imagery

Nov 21, 2019
Ritwik Gupta, Richard Hosfelt, Sandra Sajeev, Nirav Patel, Bryce Goodman, Jigar Doshi, Eric Heim, Howie Choset, Matthew Gaston

Figure 1 for xBD: A Dataset for Assessing Building Damage from Satellite Imagery

Figure 2 for xBD: A Dataset for Assessing Building Damage from Satellite Imagery

Figure 3 for xBD: A Dataset for Assessing Building Damage from Satellite Imagery

Figure 4 for xBD: A Dataset for Assessing Building Damage from Satellite Imagery

* 9 pages, 10 figures

Access Paper or Ask Questions

Constrained Generative Adversarial Networks for Interactive Image Generation

Apr 03, 2019
Eric Heim

Figure 1 for Constrained Generative Adversarial Networks for Interactive Image Generation

Figure 2 for Constrained Generative Adversarial Networks for Interactive Image Generation

Figure 3 for Constrained Generative Adversarial Networks for Interactive Image Generation

Figure 4 for Constrained Generative Adversarial Networks for Interactive Image Generation

* To Appear in the Proceedings of the 2019 Conference on Computer Vision and Pattern Recognition

Access Paper or Ask Questions

Exploiting Class Learnability in Noisy Data

Nov 15, 2018
Matthew Klawonn, Eric Heim, James Hendler

Figure 1 for Exploiting Class Learnability in Noisy Data

Figure 2 for Exploiting Class Learnability in Noisy Data

Figure 3 for Exploiting Class Learnability in Noisy Data

Figure 4 for Exploiting Class Learnability in Noisy Data

* Accepted to AAAI 2019

Access Paper or Ask Questions

Generating Triples with Adversarial Networks for Scene Graph Construction

Feb 07, 2018
Matthew Klawonn, Eric Heim

Figure 1 for Generating Triples with Adversarial Networks for Scene Graph Construction

Figure 2 for Generating Triples with Adversarial Networks for Scene Graph Construction

Figure 3 for Generating Triples with Adversarial Networks for Scene Graph Construction

Figure 4 for Generating Triples with Adversarial Networks for Scene Graph Construction

* Accepted to AAAI 2018

Access Paper or Ask Questions

Clickstream analysis for crowd-based object segmentation with confidence

Nov 29, 2017
Eric Heim, Alexander Seitel, Jonas Andrulis, Fabian Isensee, Christian Stock, Tobias Ross, Lena Maier-Hein

Figure 1 for Clickstream analysis for crowd-based object segmentation with confidence

Figure 2 for Clickstream analysis for crowd-based object segmentation with confidence

Figure 3 for Clickstream analysis for crowd-based object segmentation with confidence

Figure 4 for Clickstream analysis for crowd-based object segmentation with confidence

* to appear in IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)

Access Paper or Ask Questions

Sparse Multidimensional Patient Modeling using Auxiliary Confidence Labels

Jul 28, 2015
Eric Heim, Milos Hauskrecht

Figure 1 for Sparse Multidimensional Patient Modeling using Auxiliary Confidence Labels

Figure 2 for Sparse Multidimensional Patient Modeling using Auxiliary Confidence Labels

Figure 3 for Sparse Multidimensional Patient Modeling using Auxiliary Confidence Labels

Figure 4 for Sparse Multidimensional Patient Modeling using Auxiliary Confidence Labels

* Currently under review

Access Paper or Ask Questions