30 citations · 42 across the 12 of their papers we have counts for
15 papers
Backdoors in DRL: Four Environments Focusing on In-distribution Triggers
Chace Ashcraft, Ted Staley, Josh Carney +4
Backdoor attacks, or trojans, pose a security risk by concealing undesirable behavior in deep neural network models. Open-source neural networks are downloaded from the internet da…
Investigating the Treacherous Turn in Deep Reinforcement Learning
Chace Ashcraft, Kiran Karra, Josh Carney +1
The Treacherous Turn refers to the scenario where an artificial intelligence (AI) agent subtly, and perhaps covertly, learns to perform a behavior that benefits itself but is deeme…
Detecting Dataset Bias in Medical AI: A Generalized and Modality-Agnostic Auditing Framework
Nathan Drenkow, Mitchell Pavlak, Keith Harrigian +5
Artificial Intelligence (AI) is now firmly at the center of evidence-based medicine. Despite many success stories that edge the path of AI's rise in healthcare, there are comparabl…
A Causal Framework for Aligning Image Quality Metrics and Deep Neural Network Robustness
Nathan Drenkow, Mathias Unberath
Image quality plays an important role in the performance of deep neural networks (DNNs) that have been widely shown to exhibit sensitivity to changes in imaging conditions. Convent…
Towards Virtual Clinical Trials of Radiology AI with Conditional Generative Modeling
Benjamin D. Killeen, Bohua Wan, Aditya V. Kulkarni +4
Artificial intelligence (AI) is poised to transform healthcare by enabling personalized and efficient care through data-driven insights. Although radiology is at the forefront of A…
Causality-Driven Audits of Model Robustness
Nathan Drenkow, William Paul, Chris Ribaudo +1
Robustness audits of deep neural networks (DNN) provide a means to uncover model sensitivities to the challenging real-world imaging conditions that significantly degrade DNN perfo…
From Generalization to Precision: Exploring SAM for Tool Segmentation in Surgical Environments
Kanyifeechukwu J. Oguine, Roger D. Soberanis-Mukul, Nathan Drenkow +1
Purpose: Accurate tool segmentation is essential in computer-aided procedures. However, this task conveys challenges due to artifacts' presence and the limited training data in med…
RobustCLEVR: A Benchmark and Framework for Evaluating Robustness in Object-centric Learning
Nathan Drenkow, Mathias Unberath
Object-centric representation learning offers the potential to overcome limitations of image-level representations by explicitly parsing image scenes into their constituent compone…
Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Models
Mitchell Pavlak, Nathan Drenkow, Nicholas Petrick +2
To safely deploy deep learning-based computer vision models for computer-aided detection and diagnosis, we must ensure that they are robust and reliable. Towards that goal, algorit…
Context-Adaptive Deep Neural Networks via Bridge-Mode Connectivity
Nathan Drenkow, Alvin Tan, Chace Ashcraft +1
The deployment of machine learning models in safety-critical applications comes with the expectation that such models will perform well over a range of contexts (e.g., a vision mod…
A Systematic Review of Robustness in Deep Learning for Computer Vision: Mind the gap?
Nathan Drenkow, Numair Sani, Ilya Shpitser +1
Deep neural networks for computer vision are deployed in increasingly safety-critical and socially-impactful applications, motivating the need to close the gap in model performance…
On the Sins of Image Synthesis Loss for Self-supervised Depth Estimation
Zhaoshuo Li, Nathan Drenkow, Hao Ding +5
Scene depth estimation from stereo and monocular imagery is critical for extracting 3D information for downstream tasks such as scene understanding. Recently, learning-based method…
Patch Attack Invariance: How Sensitive are Patch Attacks to 3D Pose?
Max Lennon, Nathan Drenkow, Philippe Burlina
Perturbation-based attacks, while not physically realizable, have been the main emphasis of adversarial machine learning (ML) research. Patch-based attacks by contrast are physical…
Attack Agnostic Detection of Adversarial Examples via Random Subspace Analysis
Nathan Drenkow, Neil Fendley, Philippe Burlina
Whilst adversarial attack detection has received considerable attention, it remains a fundamentally challenging problem from two perspectives. First, while threat models can be wel…
Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformers
Zhaoshuo Li, Xingtong Liu, Nathan Drenkow +4
Stereo depth estimation relies on optimal correspondence matching between pixels on epipolar lines in the left and right images to infer depth. In this work, we revisit the problem…