activity
20202022
most citedPAC Confidence Sets for Deep Neural Networks via Calibrated Prediction

18 citations · 39 across the 5 of their papers we have counts for

collaborators

7 papers

cs.CV20223 cited

Towards PAC Multi-Object Detection and Tracking

Shuo Li, Sangdon Park, Xiayan Ji +2

Accurately detecting and tracking multi-objects is important for safety-critical applications such as autonomous navigation. However, it remains challenging to provide guarantees o…

cs.LG20224 cited

iDECODe: In-distribution Equivariance for Conformal Out-of-distribution Detection

Ramneet Kaur, Susmit Jha, Anirban Roy +4

Machine learning methods such as deep neural networks (DNNs), despite their success across different domains, are known to often generate incorrect predictions with high confidence…

cs.LG20218 cited

Detecting OODs as datapoints with High Uncertainty

Ramneet Kaur, Susmit Jha, Anirban Roy +3

Deep neural networks (DNNs) are known to produce incorrect predictions with very high confidence on out-of-distribution inputs (OODs). This limitation is one of the key challenges…

cs.LG20206 cited

PAC Confidence Predictions for Deep Neural Network Classifiers

Sangdon Park, Shuo Li, Insup Lee +1

A key challenge for deploying deep neural networks (DNNs) in safety critical settings is the need to provide rigorous ways to quantify their uncertainty. In this paper, we propose…

eess.SP2020

Joint Orthogonal Band and Power Allocation for Energy Fairness in WPT System with Nonlinear Logarithmic Energy Harvesting Model

Jaeseob Han, Gyeong Ho Lee, Sangdon Park +1

Wireless power transmission (WPT) is expected to play an important role in the Internet of Things services by providing the perpetual operation of IoT sensors. However, to prolong…

cs.LG2020

Calibrated Prediction with Covariate Shift via Unsupervised Domain Adaptation

Sangdon Park, Osbert Bastani, James Weimer +1

Reliable uncertainty estimates are an important tool for helping autonomous agents or human decision makers understand and leverage predictive models. However, existing approaches…