Global Human-guided Counterfactual Explanations for Molecular Properties via Reinforcement Learning
Authors: 
Danqing Wang, Antonis Antoniades, Kha-Dinh Luong, Edwin Zhang, Mert Kosan, Jiachen Li, Ambuj Singh, William Yang Wang, Lei LiAuthors Info & Claims
Danqing Wang, Antonis Antoniades, Kha-Dinh Luong, Edwin Zhang, Mert Kosan, Jiachen Li, Ambuj Singh, William Yang Wang, Lei LiAuthors Info & ClaimsPages 2991 - 3000
Abstract
Counterfactual explanations of Graph Neural Networks (GNNs) offer a powerful way to understand data that can naturally be represented by a graph structure. Furthermore, in many domains, it is highly desirable to derive data-driven global explanations or rules that can better explain the high-level properties of the models and data in question. However, evaluating global counterfactual explanations is hard in real-world datasets due to a lack of human-annotated ground truth, which limits their use in areas like molecular sciences. Additionally, the increasing scale of these datasets provides a challenge for random search-based methods. In this paper, we develop a novel global explanation model RLHEX for molecular property prediction. It aligns the counterfactual explanations with human-defined principles, making the explanations more interpretable and easy for experts to evaluate. RLHEX includes a VAE-based graph generator to generate global explanations and an adapter to adjust the latent representation space to human-defined principles. Optimized by Proximal Policy Optimization (PPO), the global explanations produced by RLHEX cover 4.12% more input graphs and reduce the distance between the counterfactual explanation set and the input set by 0.47% on average across three molecular datasets. RLHEX provides a flexible framework to incorporate different human-designed principles into the counterfactual explanation generation process, aligning these explanations with domain expertise. The code and data are released at https://github.com/dqwang122/RLHEX.
Supplemental Material
MOV File - RLHEX_promo
RLHEX is an end-to-end framework for generating global counterfactual explanations across large molecular datasets.
- Download
- 75.54 MB
References
[1]
Carlo Abrate and Francesco Bonchi. 2021. Counterfactual graphs for explainable classification of brain networks. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 2495--2504.
[2]
Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, et al. 2022. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862 (2022).
[3]
Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, et al. 2022. Constitutional ai: Harmlessness from ai feedback. arXiv preprint arXiv:2212.08073 (2022).
[4]
Mohit Bajaj, Lingyang Chu, Zi Yu Xue, Jian Pei, Lanjun Wang, Peter Cho-Ho Lam, and Yong Zhang. 2021. Robust counterfactual explanations on graph neural networks. Advances in Neural Information Processing Systems 34 (2021), 5644--5655.
[5]
G Richard Bickerton, Gaia V Paolini, Jérémy Besnard, Sorel Muresan, and Andrew L Hopkins. 2012. Quantifying the chemical beauty of drugs. Nature chemistry 4, 2 (2012), 90--98.
[6]
Keith T Butler, Daniel W Davies, Hugh Cartwright, Olexandr Isayev, and Aron Walsh. 2018. Machine learning for molecular and materials science. Nature 559, 7715 (2018), 547--555.
[7]
Paul Christiano, Jan Leike, Tom B. Brown, Miljan Martic, Shane Legg, and Dario Amodei. 2023. Deep reinforcement learning from human preferences. arXiv:1706.03741 [stat.ML]
[8]
Kalyan Das, Paul J Lewi, Stephen H Hughes, and Eddy Arnold. 2005. Crystallography and the design of anti-AIDS drugs: conformational flexibility and positional adaptability are important in the design of non-nucleoside HIV-1 reverse transcriptase inhibitors. Prog Biophys Mol Biol 88, 2 (Jun 2005), 209--231. https://doi.org/10.1016/j.pbiomolbio.2004.07.001
[9]
Zijie Geng, Shufang Xie, Yingce Xia, Lijun Wu, Tao Qin, Jie Wang, Yongdong Zhang, Feng Wu, and Tie-Yan Liu. 2023. De novo molecular generation via connection-aware motif mining. arXiv preprint arXiv:2302.01129 (2023).
[10]
Justin Gilmer, Samuel S Schoenholz, Patrick F Riley, Oriol Vinyals, and George E Dahl. 2017. Neural message passing for quantum chemistry. In International conference on machine learning. PMLR, 1263--1272.
[11]
Rafael Gómez-Bombarelli, Jennifer N Wei, David Duvenaud, José Miguel Hernández-Lobato, Benjamín Sánchez-Lengeling, Dennis Sheberla, Jorge Aguilera-Iparraguirre, Timothy D Hirzel, Ryan P Adams, and Alán Aspuru-Guzik. 2018. Automatic chemical design using a data-driven continuous representation of molecules. ACS central science 4, 2 (2018), 268--276.
[12]
David J. Griffiths and Darrell F. Schroeter. 2018. Introduction to Quantum Mechanics (3 ed.). Cambridge University Press.
[13]
John J Irwin, Teague Sterling, Michael M Mysinger, Erin S Bolstad, and Ryan G Coleman. 2012. ZINC: a free tool to discover chemistry for biology. Journal of chemical information and modeling 52, 7 (2012), 1757--1768.
[14]
Wengong Jin, Regina Barzilay, and Tommi Jaakkola. 2018. Junction tree variational autoencoder for molecular graph generation. In International conference on machine learning. PMLR, 2323--2332.
[15]
Jeroen Kazius, Ross McGuire, and Roberta Bursi. 2005. Derivation and validation of toxicophores for mutagenicity prediction. Journal of medicinal chemistry 48, 1 (2005), 312--320.
[16]
Diederik P. Kingma and Jimmy Ba. 2015. Adam: A Method for Stochastic Optimization. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7--9, 2015, Conference Track Proceedings, Yoshua Bengio and Yann LeCun (Eds.). http://arxiv.org/abs/1412.6980
[17]
Diederik P Kingma and Max Welling. 2013. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114 (2013).
[18]
Xiangzhe Kong, Wenbing Huang, Zhixing Tan, and Yang Liu. 2022. Molecule generation by principal subgraph mining and assembling. Advances in Neural Information Processing Systems 35 (2022), 2550--2563.
[19]
Mert Kosan, Zexi Huang, Sourav Medya, Sayan Ranu, and Ambuj Singh. 2023. Global Counterfactual Explainer for Graph Neural Networks. In WSDM.
[20]
Mert Kosan, Samidha Verma, Burouj Armgaan, Khushbu Pahwa, Ambuj Singh, Sourav Medya, and Sayan Ranu. 2024. GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking. In The Twelfth International Conference on Learning Representations. https://openreview.net/forum?id=VJvbOSXRUq
[21]
Vivian Lai, Yiming Zhang, Chacha Chen, Q Vera Liao, and Chenhao Tan. 2023. Selective explanations: Leveraging human input to align explainable ai. arXiv preprint arXiv:2301.09656 (2023).
[22]
Yujia Li, Oriol Vinyals, Chris Dyer, Razvan Pascanu, and Peter Battaglia. 2018. Learning deep generative models of graphs. arXiv preprint arXiv:1803.03324 (2018).
[23]
Tairan Liu, Misagh Naderi, Chris Alvin, Supratik Mukhopadhyay, and Michal Brylinski. 2017. Break down in order to build up: decomposing small molecules for fragment-based drug design with e molfrag. Journal of chemical information and modeling 57, 4 (2017), 627--631.
[24]
Ana Lucic, Maartje A Ter Hoeve, Gabriele Tolomei, Maarten De Rijke, and Fabrizio Silvestri. 2022. Cf-gnnexplainer: Counterfactual explanations for graph neural networks. In International Conference on Artificial Intelligence and Statistics. PMLR, 4499--4511.
[25]
Jing Ma, Ruocheng Guo, Saumitra Mishra, Aidong Zhang, and Jundong Li. 2022. Clear: Generative counterfactual explanations on graphs. Advances in Neural Information Processing Systems 35 (2022), 25895--25907.
[26]
Oscar Méndez-Lucio, Christos Nicolaou, and Berton Earnshaw. 2022. MolE: a molecular foundation model for drug discovery. arXiv:2211.02657 [q-bio.QM]
[27]
Bryan N Nguyen, Elaine W Shen, Janina Seemann, Adrienne MS Correa, James L O'Donnell, Andrew H Altieri, Nancy Knowlton, Keith A Crandall, Scott P Egan, W Owen McMillan, et al. 2020. Environmental DNA survey captures patterns of fish and invertebrate diversity across a tropical seascape. Scientific Reports 10, 1 (2020), 6729.
[28]
Danilo Numeroso and Davide Bacciu. 2021. Meg: Generating molecular counterfactual explanations for deep graph networks. In 2021 International Joint Conference on Neural Networks (IJCNN). IEEE, 1--8.
[29]
Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. 2022. Training language models to follow instructions with human feedback. arXiv:2203.02155 [cs.CL]
[30]
Florbela Pereira and João Aires-de Sousa. 2018. Machine learning for the prediction of molecular dipole moments obtained by density functional theory. Journal of cheminformatics 10 (2018), 1--11.
[31]
Mario Alfonso Prado-Romero, Bardh Prenkaj, Giovanni Stilo, and Fosca Giannotti. [n. d.]. A survey on graph counterfactual explanations: definitions, methods, evaluation, and research challenges. Comput. Surveys ([n. d.]).
[32]
Kaspar Riesen, Horst Bunke, et al . 2008. IAM Graph Database Repository for Graph Based Pattern Recognition and Machine Learning. In SSPR/SPR, Vol. 5342. 287--297.
[33]
Alexander Rives, Joshua Meier, Tom Sercu, Siddharth Goyal, Zeming Lin, Jason Liu, Demi Guo, Myle Ott, C. Lawrence Zitnick, Jerry Ma, and Rob Fergus. 2021. Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. Proceedings of the National Academy of Sciences 118, 15 (2021), e2016239118. https://doi.org/10.1073/pnas.2016239118 arXiv:https://www.pnas.org/doi/pdf/10.1073/pnas.2016239118
[34]
Jerret Ross, Brian Belgodere, Vijil Chenthamarakshan, Inkit Padhi, Youssef Mroueh, and Payel Das. 2022. Large-Scale Chemical Language Representations Capture Molecular Structure and Properties. arXiv:2106.09553 [cs.LG]
[35]
Matthias Rupp, Alexandre Tkatchenko, Klaus-Robert Müller, and O. Anatole von Lilienfeld. 2012. Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning. Phys. Rev. Lett. 108 (Jan 2012), 058301. Issue 5. https://doi.org/10.1103/PhysRevLett.108.058301
[36]
John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov. 2017. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347 (2017).
[37]
Martin Simonovsky and Nikos Komodakis. 2018. Graphvae: Towards generation of small graphs using variational autoencoders. In Artificial Neural Networks and Machine Learning--ICANN 2018: 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4--7, 2018, Proceedings, Part I 27. Springer, 412--422.
[38]
Zhiqing Sun, Yikang Shen, Hongxin Zhang, Qinhong Zhou, Zhenfang Chen, David Cox, Yiming Yang, and Chuang Gan. 2023. SALMON: Self-Alignment with Principle-Following Reward Models. arXiv:2310.05910 [cs.LG]
[39]
Zhiqing Sun, Yikang Shen, Qinhong Zhou, Hongxin Zhang, Zhenfang Chen, David Cox, Yiming Yang, and Chuang Gan. 2023. Principle-Driven Self- Alignment of Language Models from Scratch with Minimal Human Supervision. arXiv:2305.03047 [cs.LG]
[40]
Nathan J. Szymanski, Bernardus Rendy, Yuxing Fei, Rishi E. Kumar, Tanjin He, David Milsted, Matthew J. McDermott, Max Gallant, Ekin Dogus Cubuk, Amil Merchant, Haegyeom Kim, Anubhav Jain, Christopher J. Bartel, Kristin Persson, Yan Zeng, and Gerbrand Ceder. 2023. An autonomous laboratory for the accelerated synthesis of novel materials. Nature 624, 7990 (2023), 86--91. https://doi.org/10.1038/s41586-023-06734-w
[41]
Juntao Tan, Shijie Geng, Zuohui Fu, Yingqiang Ge, Shuyuan Xu, Yunqi Li, and Yongfeng Zhang. 2022. Learning and evaluating graph neural network explanations based on counterfactual and factual reasoning. In Proceedings of the ACM Web Conference 2022. 1018--1027.
[42]
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems 30 (2017).
[43]
Xiang Wang, Yingxin Wu, An Zhang, Fuli Feng, Xiangnan He, and Tat-Seng Chua. 2022. Reinforced causal explainer for graph neural networks. IEEE Transactions on Pattern Analysis and Machine Intelligence 45, 2 (2022), 2297--2309.
[44]
Zhi Wang, Chicheng Zhang, and Kamalika Chaudhuri. 2022. Thompson Sampling for Robust Transfer in Multi-Task Bandits. arXiv:2206.08556 [cs.LG]
[45]
Oliver Wieder, Stefan Kohlbacher, Mélaine Kuenemann, Arthur Garon, Pierre Ducrot, Thomas Seidel, and Thierry Langer. 2020. A compact review of molecular property prediction with graph neural networks. Drug Discovery Today: Technologies 37 (2020), 1--12.
[46]
Felix Wong, Erica J Zheng, Jacqueline A Valeri, Nina M Donghia, Melis N Anahtar, Satotaka Omori, Alicia Li, Andres Cubillos-Ruiz, Aarti Krishnan, Wengong Jin, et al. 2023. Discovery of a structural class of antibiotics with explainable deep learning. Nature (2023), 1--9.
[47]
Zhenxing Wu, Jike Wang, Hongyan Du, Dejun Jiang, Yu Kang, Dan Li, Peichen Pan, Yafeng Deng, Dongsheng Cao, Chang-Yu Hsieh, et al . 2023. Chemistry-intuitive explanation of graph neural networks for molecular property prediction with substructure masking. Nature Communications 14, 1 (2023), 2585.
[48]
Yutong Xie, Chence Shi, Hao Zhou, Yuwei Yang, Weinan Zhang, Yong Yu, and Lei Li. 2021. MARS: Markov Molecular Sampling for Multi-objective Drug Discovery. In International Conference on Learning Representations.
[49]
Wenda Xu, Danqing Wang, Liangming Pan, Zhenqiao Song, Markus Freitag, William Yang Wang, and Lei Li. 2023. Instructscore: Towards explainable text generation evaluation with automatic feedback. arXiv preprint arXiv:2305.14282 (2023).
[50]
Qiang Yang, Changsheng Ma, Qiannan Zhang, Xin Gao, Chuxu Zhang, and Xiangliang Zhang. 2023. Counterfactual Learning on Heterogeneous Graphs with Greedy Perturbation. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (Long Beach, CA, USA) (KDD '23). Association for Computing Machinery, New York, NY, USA, 2988--2998. https://doi.org/10.1145/3580305.3599289
[51]
Jiaxuan You, Bowen Liu, Zhitao Ying, Vijay Pande, and Jure Leskovec. 2018. Graph convolutional policy network for goal-directed molecular graph generation. Advances in neural information processing systems 31 (2018).
[52]
Hao Yuan, Jiliang Tang, Xia Hu, and Shuiwang Ji. 2020. Xgnn: Towards model-level explanations of graph neural networks. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 430--438.
[53]
Ziwei Zhang, Peng Cui, and Wenwu Zhu. 2020. Deep learning on graphs: A survey. IEEE Transactions on Knowledge and Data Engineering 34, 1 (2020), 249--270.
[54]
Jie Zhou, Ganqu Cui, Shengding Hu, Zhengyan Zhang, Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, and Maosong Sun. 2020. Graph neural networks: A review of methods and applications. AI open 1 (2020), 57present--81.
[55]
Daniel M. Ziegler, Nisan Stiennon, Jeffrey Wu, Tom B. Brown, Alec Radford, Dario Amodei, Paul F. Christiano, and Geoffrey Irving. 2019. Fine-Tuning Language Models from Human Preferences. CoRR abs/1909.08593 (2019). arXiv:1909.08593 http://arxiv.org/abs/1909.08593
Index Terms
- Global Human-guided Counterfactual Explanations for Molecular Properties via Reinforcement Learning
Recommendations
Counterfactual explanations for misclassified images: How human and machine explanations differ
AbstractCounterfactual explanations have emerged as a popular solution for the eXplainable AI (XAI) problem of elucidating the predictions of black-box deep-learning systems because people easily understand them, they apply across different problem ...
Highlights- A novel user-centric methodology where users create counterfactual explanations.
- Human-generated counterfactuals differ significantly from machine-generated ones.
- A novel interpretation of this difference using the notion of “...
Prototype-Guided Counterfactual Explanations via Variational Auto-encoder for Recommendation
Machine Learning and Knowledge Discovery in Databases: Applied Data Science and Demo TrackAbstractCounterfactual reasoning has recently achieved impressive performance in the explainability of recommendation. However, existing counterfactual explainable methods ignore the realism of explanations and consider only the sparsity and proximity of ...
Counterfactual Explanations for Reinforcement Learning Agents
AAMAS '23: Proceedings of the 2023 International Conference on Autonomous Agents and Multiagent SystemsReinforcement learning (RL) algorithms often use neural networks to represent agent's policy, making them difficult to interpret. Counterfactual explanations are human-friendly explanations which offer users actionable advice on how to change their ...
Comments
Information & Contributors
Information
Published In
August 2024
6901 pages
Copyright © 2024 Owner/Author.
This work is licensed under a Creative Commons Attribution International 4.0 License.
Sponsors
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 24 August 2024
Check for updates
Author Tags
Qualifiers
- Research-article
Funding Sources
- National Science Foundation
Conference
KDD '24: The 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
August 25 - 29, 2024
Barcelona, Spain
Acceptance Rates
Overall Acceptance Rate 1,133 of 8,635 submissions, 13%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 79Total Downloads
- Downloads (Last 12 months)79
- Downloads (Last 6 weeks)79
Reflects downloads up to 23 Sep 2024
Other Metrics
Citations
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in