RI: Small: Enabling Interpretable AI via Bayesian Deep Learning

RI：小型：通过贝叶斯深度学习实现可解释的人工智能

• 批准号: 2127918
• 负责人: Hao Wang
• 金额: $ 49.99万
• 依托单位: Rutgers University New Brunswick
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2127918&HistoricalAwards=false
• 关键词: RI; Small; Enabling; Interpretable; AI

Interpretability is one of the fundamental obstacles on the adoption and deployment of deep-learning-based AI systems across various fields such as healthcare, e-commerce, transportation, earth science, and manufacturing. An ideal interpretable model should be able to interpret its prediction using human-understandable concepts (e.g., “color” and “shape”), conform to conditional dependencies in the real world (e.g., whether a customer's purchase is due to a discount), and handle uncertainty in data (e.g., how certain the model is about the rainfall tomorrow). Unfortunately, deep learning as a connectionist approach does not natively support these desiderata. The goal of this project is to develop a general interpreter framework for deep learning models. Interpreters under this framework can be plugged into a deep learning model and interpret its predictions using a graph of human-understandable concepts, without sacrificing the model’s performance. Methods developed in this project will be applied in health monitoring to interpret models’ reasoning on patient status, and in recommender systems to interpret models’ recommended items for users.This project will develop two sets of methods based on Bayesian deep learning: (1) “Bayesian deep interpreters” that interpret deep learning models with graphical models describing the conditional dependencies leading to current predictions. (2) “Bayesian deep controllers” that control deep learning models' predictions by manipulating specific random variables in the graphical models attached to the controlled models. Development of such novel methods will build intellectual and formal connection between deep learning and probabilistic graphical models, two major machine learning paradigms that have long been seen as incompatible. It will advance the state of the art on machine learning and AI by: (1) formulating a new Bayesian deep learning framework to unify deep learning and graphical models, the synergy of which will significantly improve deep learning interpretability, (2) under such a principled framework, designing concrete methods that are plug-and-play and therefore do not sacrifice the deep learning models' performance (e.g., accuracy), (3) investigating what theoretical guarantees the developed methods provide and therefore laying foundations for future work by the team and the community, (4) analyzing the trade-off between accuracy, interpretability, and controllability and providing design guidance for interpretable AI systems.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.

可解释性是在医疗保健，电子商务，运输，地球科学和制造等各个领域采用和部署基于深度学习的AI系统的基本障碍之一。理想的可解释模型应该能够使用人类理解的概念（例如“颜色”和“形状”）来解释其预测，符合现实世界中有条件的依赖性（例如，客户的购买是由于折扣而造成的），并处理数据的不确定性（例如，某些模型如何明天降雨）。不幸的是，作为连接主义方法的深度学习并不能本地支持这些需求。该项目的目的是为深度学习模型开发一个通用解释框架。在此框架下的口译员可以插入深度学习模型中，并使用人类可行的概念图来解释其预测，而无需牺牲模型的性能。该项目中开发的方法将应用于健康监控，以解释模型对患者状态的推理，并在推荐系统中解释模型的用户推荐项目。该项目将基于贝叶斯深度学习开发两组方法：（1）“贝叶斯深度解释者”，“贝叶斯深度解释者”以图形模型来解释深度学习模型，以描述有条件依赖的图形模型。 （2）控制深度学习模型的“贝叶斯深控制器”，通过操纵与受控模型相关的图形模型中的特定随机变量来控制深度学习模型的预测。这种新颖方法的开发将在深度学习和概率图形模型之间建立智力和正式联系，这是两个主要的机器学习范式，这些范式长期以来一直被视为不相容。它将通过：（1）制定一个新的贝叶斯深度学习框架来统一深度学习和图形模型，其协同作用将显着改善深度学习解释性，（2）在此类主要的框架下，设计是插件和开发的方法，（2）提供了发展模型的效果（E.G. E.G. 3），（3），（2）因此，（4）分析准确性，可解释性和可控性之间的权衡，并为可解释的AI系统提供设计指南。该奖项反映了NSF的法定任务，并被认为是通过使用基金会的知识分子优点和更广泛的影响审查标准来通过评估来获得的支持。

项目成果

System 1 + System 2 = Better World: Neural-Symbolic Chain of Logic Reasoning

• DOI: 10.18653/v1/2022.findings-emnlp.42
• 发表时间: 2022
• 作者: Wenyue Hua;Yongfeng Zhang

“My nose is running.” “Are you also coughing?”: Building a medical diagnosis agent with interpretable inquiry logics

“我流鼻涕了。”“你也在咳嗽吗？”：构建具有可解释查询逻辑的医疗诊断代理

• DOI: 10.24963/ijcai.2022/592
• 发表时间: 2022
• 期刊: Proceedings of the International Conferences on Artificial Intelligence
• 作者: Liu, Wenge;Cheng, Yi;Wang, Hao;Tang, Jianheng;Liu, Yafei;Zhao, Ruihui;Li, Wenjie;Zheng, Yefeng;Liang, Xiaodan
• 通讯作者: Liang, Xiaodan

Deconfounded Causal Collaborative Filtering

• DOI: 10.1145/3606035
• 发表时间: 2021-10
• 期刊: ACM Transactions on Recommender Systems
• 作者: Shuyuan Xu;Juntao Tan;Shelby Heinecke;Jia Li;Yongfeng Zhang

Learn Basic Skills and Reuse: Modularized Adaptive Neural Architecture Search (MANAS)

• DOI: 10.1145/3511808.3557385
• 发表时间: 2022-08
• 期刊: Proceedings of the 31st ACM International Conference on Information & Knowledge Management
• 作者: H. Chen;Yunqi Li;He Zhu;Yongfeng Zhang

Causal Collaborative Filtering

• DOI: 10.1145/3578337.3605122
• 发表时间: 2021-02
• 期刊: Proceedings of the 2023 ACM SIGIR International Conference on Theory of Information Retrieval
• 作者: Shuyuan Xu;Yingqiang Ge;Yunqi Li;Zuohui Fu;Xu Chen;Yongfeng Zhang