-
Clever Hans in Chemistry: Chemist Style Signals Confound Activity Prediction on Public Benchmarks
Authors:
Andrew D. Blevins,
Ian K. Quigley
Abstract:
Can machine learning models identify which chemist made a molecule from structure alone? If so, models trained on literature data may exploit chemist intent rather than learning causal structure-activity relationships. We test this by linking CHEMBL assays to publication authors and training a 1,815-class classifier to predict authors from molecular fingerprints, achieving 60% top-5 accuracy under…
▽ More
Can machine learning models identify which chemist made a molecule from structure alone? If so, models trained on literature data may exploit chemist intent rather than learning causal structure-activity relationships. We test this by linking CHEMBL assays to publication authors and training a 1,815-class classifier to predict authors from molecular fingerprints, achieving 60% top-5 accuracy under scaffold-based splitting. We then train an activity model that receives only a protein identifier and an author-probability vector derived from structure, with no direct access to molecular descriptors. This author-only model achieves predictive power comparable to a simple baseline that has access to structure. This reveals a "Clever Hans" failure mode: models can predict bioactivity largely by inferring chemist goals and favorite targets without requiring a lab-independent understanding of chemistry. We analyze the sources of this leakage, propose author-disjoint splits, and recommend dataset practices to decouple chemist intent from biological outcomes.
△ Less
Submitted 23 December, 2025;
originally announced December 2025.
-
Higher-Order Equivariant Neural Networks for Charge Density Prediction in Materials
Authors:
Teddy Koker,
Keegan Quigley,
Eric Taw,
Kevin Tibbetts,
Lin Li
Abstract:
The calculation of electron density distribution using density functional theory (DFT) in materials and molecules is central to the study of their quantum and macro-scale properties, yet accurate and efficient calculation remains a long-standing challenge. We introduce ChargE3Net, an E(3)-equivariant graph neural network for predicting electron density in atomic systems. ChargE3Net enables the lea…
▽ More
The calculation of electron density distribution using density functional theory (DFT) in materials and molecules is central to the study of their quantum and macro-scale properties, yet accurate and efficient calculation remains a long-standing challenge. We introduce ChargE3Net, an E(3)-equivariant graph neural network for predicting electron density in atomic systems. ChargE3Net enables the learning of higher-order equivariant feature to achieve high predictive accuracy and model expressivity. We show that ChargE3Net exceeds the performance of prior work on diverse sets of molecules and materials. When trained on the massive dataset of over 100K materials in the Materials Project database, our model is able to capture the complexity and variability in the data, leading to a significant 26.7% reduction in self-consistent iterations when used to initialize DFT calculations on unseen materials. Furthermore, we show that non-self-consistent DFT calculations using our predicted charge densities yield near-DFT performance on electronic and thermodynamic property prediction at a fraction of the computational cost. Further analysis attributes the greater predictive accuracy to improved modeling of systems with high angular variations. These results illuminate a pathway towards a machine learning-accelerated ab initio calculations for materials discovery.
△ Less
Submitted 14 May, 2024; v1 submitted 8 December, 2023;
originally announced December 2023.
-
Improving Medical Visual Representations via Radiology Report Generation
Authors:
Keegan Quigley,
Miriam Cha,
Josh Barua,
Geeticka Chauhan,
Seth Berkowitz,
Steven Horng,
Polina Golland
Abstract:
Vision-language pretraining has been shown to produce high-quality visual encoders which transfer efficiently to downstream computer vision tasks. Contrastive learning approaches have increasingly been adopted for medical vision language pretraining (MVLP), yet recent developments in generative AI offer new modeling alternatives. This paper introduces RadTex, a CNN-encoder transformer-decoder arch…
▽ More
Vision-language pretraining has been shown to produce high-quality visual encoders which transfer efficiently to downstream computer vision tasks. Contrastive learning approaches have increasingly been adopted for medical vision language pretraining (MVLP), yet recent developments in generative AI offer new modeling alternatives. This paper introduces RadTex, a CNN-encoder transformer-decoder architecture optimized for radiology. We explore bidirectional captioning as an alternative MVLP strategy and demonstrate that RadTex's captioning pretraining is competitive with established contrastive methods, achieving a CheXpert macro-AUC of 89.4%. Additionally, RadTex's lightweight text decoder not only generates clinically relevant radiology reports (macro-F1 score of 0.349), but also provides targeted, interactive responses, highlighting the utility of bidirectional captioning in advancing medical image analysis.
△ Less
Submitted 10 January, 2025; v1 submitted 30 October, 2023;
originally announced October 2023.
-
Graph Contrastive Learning for Materials
Authors:
Teddy Koker,
Keegan Quigley,
Will Spaeth,
Nathan C. Frey,
Lin Li
Abstract:
Recent work has shown the potential of graph neural networks to efficiently predict material properties, enabling high-throughput screening of materials. Training these models, however, often requires large quantities of labelled data, obtained via costly methods such as ab initio calculations or experimental evaluation. By leveraging a series of material-specific transformations, we introduce Cry…
▽ More
Recent work has shown the potential of graph neural networks to efficiently predict material properties, enabling high-throughput screening of materials. Training these models, however, often requires large quantities of labelled data, obtained via costly methods such as ab initio calculations or experimental evaluation. By leveraging a series of material-specific transformations, we introduce CrystalCLR, a framework for constrastive learning of representations with crystal graph neural networks. With the addition of a novel loss function, our framework is able to learn representations competitive with engineered fingerprinting methods. We also demonstrate that via model finetuning, contrastive pretraining can improve the performance of graph neural networks for prediction of material properties and significantly outperform traditional ML models that use engineered fingerprints. Lastly, we observe that CrystalCLR produces material representations that form clusters by compound class.
△ Less
Submitted 23 November, 2022;
originally announced November 2022.
-
Developing Modular Autonomous Capabilities for sUAS Operations
Authors:
Keegan Quigley,
Virginia Goodwin,
Luis Alvarez,
Justin Yao,
Yousef Salaman Maclara
Abstract:
Small teams in the field can benefit from the capabilities provided by small Uncrewed Aerial Systems (sUAS) for missions such as reconnaissance, hostile attribution, remote emplacement, and search and rescue. The mobility, communications, and flexible payload capacity of sUAS can offer teams new levels of situational awareness and enable more highly coordinated missions than previously possible. H…
▽ More
Small teams in the field can benefit from the capabilities provided by small Uncrewed Aerial Systems (sUAS) for missions such as reconnaissance, hostile attribution, remote emplacement, and search and rescue. The mobility, communications, and flexible payload capacity of sUAS can offer teams new levels of situational awareness and enable more highly coordinated missions than previously possible. However, piloting such aircraft for specific missions draws personnel away from other mission-critical tasks, increasing the load on remaining personnel while also increasing complexity of operations. For wider adoption and use of sUAS for security and humanitarian missions, safe and robust autonomy must be employed to reduce this burden on small teams. In this paper, we present the development of the Collaborative-UAS for Hostile Attribution, Surveillance, Emplacement, and Reconnaissance (CHASER) testbed, for rapidly prototyping capabilities that will reduce strain on small teams through sensor-guided autonomous control. We attempt to address autonomy needs unfilled by commercial sUAS platforms by creating and testing a series of composable modules that can be configured to support multiple missions. Methods implemented and presented here include radar track correlation, on-board computer vision target detection, target position estimation, closed-loop relative position control, and efficient search of a 3D volume for target acquisition. We configure and test a series of these modules in an example mission, executing a fully autonomous chase of an intruding sUAS in live flight, and demonstrating the success of the modularized autonomy approach. We present performance results from simulation or live flight tests for each module. Lastly, we describe the software architecture that we have developed for flexible controls and comment on how the capabilities presented may enable additional missions.
△ Less
Submitted 27 January, 2023; v1 submitted 1 November, 2022;
originally announced November 2022.
-
RadTex: Learning Efficient Radiograph Representations from Text Reports
Authors:
Keegan Quigley,
Miriam Cha,
Ruizhi Liao,
Geeticka Chauhan,
Steven Horng,
Seth Berkowitz,
Polina Golland
Abstract:
Automated analysis of chest radiography using deep learning has tremendous potential to enhance the clinical diagnosis of diseases in patients. However, deep learning models typically require large amounts of annotated data to achieve high performance -- often an obstacle to medical domain adaptation. In this paper, we build a data-efficient learning framework that utilizes radiology reports to im…
▽ More
Automated analysis of chest radiography using deep learning has tremendous potential to enhance the clinical diagnosis of diseases in patients. However, deep learning models typically require large amounts of annotated data to achieve high performance -- often an obstacle to medical domain adaptation. In this paper, we build a data-efficient learning framework that utilizes radiology reports to improve medical image classification performance with limited labeled data (fewer than 1000 examples). Specifically, we examine image-captioning pretraining to learn high-quality medical image representations that train on fewer examples. Following joint pretraining of a convolutional encoder and transformer decoder, we transfer the learned encoder to various classification tasks. Averaged over 9 pathologies, we find that our model achieves higher classification performance than ImageNet-supervised and in-domain supervised pretraining when labeled training data is limited.
△ Less
Submitted 7 April, 2023; v1 submitted 5 August, 2022;
originally announced August 2022.
-
Multimodal Representation Learning via Maximization of Local Mutual Information
Authors:
Ruizhi Liao,
Daniel Moyer,
Miriam Cha,
Keegan Quigley,
Seth Berkowitz,
Steven Horng,
Polina Golland,
William M. Wells
Abstract:
We propose and demonstrate a representation learning approach by maximizing the mutual information between local features of images and text. The goal of this approach is to learn useful image representations by taking advantage of the rich information contained in the free text that describes the findings in the image. Our method trains image and text encoders by encouraging the resulting represe…
▽ More
We propose and demonstrate a representation learning approach by maximizing the mutual information between local features of images and text. The goal of this approach is to learn useful image representations by taking advantage of the rich information contained in the free text that describes the findings in the image. Our method trains image and text encoders by encouraging the resulting representations to exhibit high local mutual information. We make use of recent advances in mutual information estimation with neural network discriminators. We argue that the sum of local mutual information is typically a lower bound on the global mutual information. Our experimental results in the downstream image classification tasks demonstrate the advantages of using local features for image-text representation learning.
△ Less
Submitted 14 December, 2021; v1 submitted 7 March, 2021;
originally announced March 2021.
-
Accessibility of Virtual Reality Locomotion Modalities to Adults and Minors
Authors:
Zhijiong Huang,
Yu Zhang,
Kathryn C. Quigley,
Ramya Sankar,
Clemence Wormser,
Xinxin Mo,
Allen Y. Yang
Abstract:
Virtual reality (VR) is an important new technology that is fun-damentally changing the way people experience entertainment and education content. Due to the fact that most currently available VR products are one size fits all, the accessibility of the content design and user interface design, even for healthy children is not well understood. It requires more research to ensure that children can h…
▽ More
Virtual reality (VR) is an important new technology that is fun-damentally changing the way people experience entertainment and education content. Due to the fact that most currently available VR products are one size fits all, the accessibility of the content design and user interface design, even for healthy children is not well understood. It requires more research to ensure that children can have equally good user compared to adults in VR. In our study, we seek to explore accessibility of locomotion in VR between healthy adults and minors along both objective and subjective dimensions. We performed a user experience experiment where subjects completed a simple task of moving and touching underwater animals in VR using one of four different locomotion modalities, as well as real-world walking without wearing VR headsets as the baseline. Our results show that physical body movement that mirrors real-world movement exclusively is the least preferred by both adults and minors. However, within the different modalities of controller assisted locomotion there are variations between adults and minors for preference and challenge levels.
△ Less
Submitted 16 April, 2019;
originally announced April 2019.