Collaborative Research: SCH: Geometry and Topology for Interpretable and Reliable Deep Learning in Medical Imaging

合作研究：SCH：医学成像中可解释且可靠的深度学习的几何和拓扑

• 批准号: 2205418
• 负责人: Bei Phillips
• 金额: $ 57.01万
• 依托单位: University of Utah
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-01 至 2026-08-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2205418&HistoricalAwards=false
• 关键词: Collaborative; Research; SCH; Geometry; Topology

Deep learning models are being developed for safety-critical applications, such as health care, autonomous vehicles, and security. Their impressive performance has the potential to make profound impacts on human lives. For example, deep neural networks (DNNs) in medical imaging have been shown to have impressive diagnostic capabilities, often near that of expert radiologists. However, deep learning has not made it into standard clinical care, primarily due to a lack of understanding of why a model works and why it fails. The goal of this project is to develop methods for making machine learning models interpretable and reliable, and thus bridge the trust gap to make machine learning translatable to the clinic. This project achieves this goal through investigation of the mathematical foundations -- specifically the geometry and topology -- of DNNs. Based on these mathematical foundations, this project will develop computational tools that will improve the interpretability and reliability of DNNs. The methods developed in this project will be broadly applicable wherever deep learning is used, including health care, security, computer vision, natural language processing, etc.The power of a deep neural network lies in its hidden layers, where the network learns internal representations of input data. This research project centers around the hypothesis that geometry and topology provide critical tools for analyzing the internal representations of DNNs. The first goal of this project is to develop a rigorous mathematical and algorithmic foundation for describing the geometry and topology of a neural network's internal representations and then design efficient algorithms for geometric and topological computations necessary to explore these spaces. The next aim of this project is to apply these tools to improve the interpretability of deep learning. This will be done by linking a model's internal representation with interpretable and trusted features and by interactive visualization that explores the landscape of a model's internal representation. The next goal of this project focuses on model reliability, where geometry and topology will be used for failure identification, mitigation, and prevention. Finally, this project will test the developed techniques for reliable and interpretable neural networks in a real-world setting to aid expert oncologists in predicting patient outcomes in head and neck cancers, e.g., whether a tumor will metastasize.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

正在为医疗保健、自动驾驶汽车和安全等安全关键应用开发深度学习模型。他们令人印象深刻的表现有可能对人类生活产生深远的影响。例如，医学成像中的深度神经网络 (DNN) 已被证明具有令人印象深刻的诊断能力，通常接近专家放射科医生的诊断能力。然而，深度学习尚未成为标准的临床护理，主要是由于缺乏对模型为何有效以及为何失败的理解。该项目的目标是开发使机器学习模型可解释且可靠的方法，从而弥合信任差距，使机器学习可转化为临床。该项目通过研究 DNN 的数学基础（特别是几何和拓扑）来实现这一目标。基于这些数学基础，该项目将开发计算工具，以提高 DNN 的可解释性和可靠性。该项目开发的方法将广泛适用于使用深度学习的任何地方，包括医疗保健、安全、计算机视觉、自然语言处理等。深度神经网络的力量在于其隐藏层，网络在隐藏层中学习内部表示输入数据。该研究项目围绕这样一个假设：几何和拓扑为分析 DNN 的内部表示提供了关键工具。该项目的第一个目标是开发严格的数学和算法基础来描述神经网络内部表示的几何和拓扑，然后设计探索这些空间所需的几何和拓扑计算的有效算法。该项目的下一个目标是应用这些工具来提高深度学习的可解释性。这将通过将模型的内部表示与可解释和可信的特征联系起来，并通过交互式可视化来探索模型内部表示的景观来完成。该项目的下一个目标侧重于模型可靠性，其中几何和拓扑将用于故障识别、缓解和预防。最后，该项目将在现实环境中测试所开发的可靠且可解释的神经网络技术，以帮助肿瘤学家预测头颈癌患者的结果，例如肿瘤是否会转移。该奖项反映了 NSF 的法定使命和通过使用基金会的智力价值和更广泛的影响审查标准进行评估，该项目被认为值得支持。

项目成果

TopoBERT: Exploring the topology of fine-tuned word representations

TopoBERT：探索微调词表示的拓扑

• DOI: 10.1177/14738716231168671
• 发表时间: 2023-07
• 期刊: Information Visualization
• 影响因子: 2.3
• 作者: Rathore, Archit;Zhou, Yichu;Srikumar, Vivek;Wang, Bei
• 通讯作者: Wang, Bei

VERB: Visualizing and Interpreting Bias Mitigation Techniques Geometrically for Word Representations

VERB：以几何方式可视化和解释单词表示的偏差缓解技术

• 发表时间: 2023-06
• 期刊: ACM Transactions on Interactive Intelligent Systems
• 影响因子: 3.4
• 作者: Rathore, Archit;Dev, Sunipa;Phillips, Jeff M.;Srikumar, Vivek;Zheng, Yan;Yeh, Chin;Wang, Junpeng;Zhang, Wei;Wang, Bei
• 通讯作者: Wang, Bei

Visualizing and Analyzing the Topology of Neuron Activations in Deep Adversarial Training.

可视化和分析深度对抗训练中神经元激活的拓扑。

• 发表时间: 2023-07
• 期刊: and Geometry in Machine Learning (TAGML
• 作者: Zhou, Youjia;Zhou, Yi;Ding, Jie;Wang, Bei.
• 通讯作者: Wang, Bei.

Experimental Observations of the Topology of Convolutional Neural Network Activations

卷积神经网络激活拓扑的实验观察

• 发表时间: 2023-06
• 期刊: Proceedings of the AAAI Conference on Artificial Intelligence
• 作者: Purvine, Emilie;Brown, Davis;Jefferson, Brett;Joslyn, Cliff;Praggastis, Brenda;Rathore, Archit;Shapiro, Madelyn;Wang, Bei;Zhou, Youjia
• 通讯作者: Zhou, Youjia