A key impediment to reinforcement learning (RL) in real applications with limited, batch data is in defining a reward function that reflects what we implicitly know about reasonable behaviour for a task and allows for robust off-policy evaluation. In this work, we develop a method to identify an admissible set of reward functions for policies that (a) do not deviate too far in performance from prior behaviour, and (b) can be evaluated with high confidence, given only a collection of past trajectories. Together, these ensure that we avoid proposing unreasonable policies in high-risk settings. We demonstrate our approach to reward design on synthetic domains as well as in a critical care context, to guide the design of a reward function that consolidates clinical objectives to learn a policy for weaning patients from mechanical ventilation.
{"title":"Defining admissible rewards for high-confidence policy evaluation in batch reinforcement learning","authors":"Niranjani Prasad, B. Engelhardt, F. Doshi-Velez","doi":"10.1145/3368555.3384450","DOIUrl":"https://doi.org/10.1145/3368555.3384450","url":null,"abstract":"A key impediment to reinforcement learning (RL) in real applications with limited, batch data is in defining a reward function that reflects what we implicitly know about reasonable behaviour for a task and allows for robust off-policy evaluation. In this work, we develop a method to identify an admissible set of reward functions for policies that (a) do not deviate too far in performance from prior behaviour, and (b) can be evaluated with high confidence, given only a collection of past trajectories. Together, these ensure that we avoid proposing unreasonable policies in high-risk settings. We demonstrate our approach to reward design on synthetic domains as well as in a critical care context, to guide the design of a reward function that consolidates clinical objectives to learn a policy for weaning patients from mechanical ventilation.","PeriodicalId":87342,"journal":{"name":"Proceedings of the ACM Conference on Health, Inference, and Learning","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76472265","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chirag Nagpal, Dennis Wei, B. Vinzamuri, Monica Shekhar, Sara E. Berger, Subhro Das, Kush R. Varshney
The dearth of prescribing guidelines for physicians is one key driver of the current opioid epidemic in the United States. In this work, we analyze medical and pharmaceutical claims data to draw insights on characteristics of patients who are more prone to adverse outcomes after an initial synthetic opioid prescription. Toward this end, we propose a generative model that allows discovery from observational data of subgroups that demonstrate an enhanced or diminished causal effect due to treatment. Our approach models these sub-populations as a mixture distribution, using sparsity to enhance interpretability, while jointly learning nonlinear predictors of the potential outcomes to better adjust for confounding. The approach leads to human interpretable insights on discovered subgroups, improving the practical utility for decision support.
{"title":"Interpretable subgroup discovery in treatment effect estimation with application to opioid prescribing guidelines","authors":"Chirag Nagpal, Dennis Wei, B. Vinzamuri, Monica Shekhar, Sara E. Berger, Subhro Das, Kush R. Varshney","doi":"10.1145/3368555.3384456","DOIUrl":"https://doi.org/10.1145/3368555.3384456","url":null,"abstract":"The dearth of prescribing guidelines for physicians is one key driver of the current opioid epidemic in the United States. In this work, we analyze medical and pharmaceutical claims data to draw insights on characteristics of patients who are more prone to adverse outcomes after an initial synthetic opioid prescription. Toward this end, we propose a generative model that allows discovery from observational data of subgroups that demonstrate an enhanced or diminished causal effect due to treatment. Our approach models these sub-populations as a mixture distribution, using sparsity to enhance interpretability, while jointly learning nonlinear predictors of the potential outcomes to better adjust for confounding. The approach leads to human interpretable insights on discovered subgroups, improving the practical utility for decision support.","PeriodicalId":87342,"journal":{"name":"Proceedings of the ACM Conference on Health, Inference, and Learning","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75249141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Kathleen M. Lewis, Guha Balakrishnan, N. Rost, J. Guttag, Adrian V. Dalca
We introduce SparseVM, a method that registers clinical-quality 3D MR scans both faster and more accurately than previously possible. Deformable alignment, or registration, of clinical scans is a fundamental task for many clinical neuroscience studies. However, most registration algorithms are designed for high-resolution research-quality scans. In contrast to research-quality scans, clinical scans are often sparse, missing up to 86% of the slices available in research-quality scans. Existing methods for registering these sparse images are either inaccurate or extremely slow. We present a learning-based registration method, SparseVM, that is more accurate and orders of magnitude faster than the most accurate clinical registration methods. To our knowledge, it is the first method to use deep learning specifically tailored to registering clinical images. We demonstrate our method on a clinically-acquired MRI dataset of stroke patients and on a simulated sparse MRI dataset. Our code is available as part of the VoxelMorph package at http://voxelmorph.mit.edu.
{"title":"Fast learning-based registration of sparse 3D clinical images","authors":"Kathleen M. Lewis, Guha Balakrishnan, N. Rost, J. Guttag, Adrian V. Dalca","doi":"10.1145/3368555.3384462","DOIUrl":"https://doi.org/10.1145/3368555.3384462","url":null,"abstract":"We introduce SparseVM, a method that registers clinical-quality 3D MR scans both faster and more accurately than previously possible. Deformable alignment, or registration, of clinical scans is a fundamental task for many clinical neuroscience studies. However, most registration algorithms are designed for high-resolution research-quality scans. In contrast to research-quality scans, clinical scans are often sparse, missing up to 86% of the slices available in research-quality scans. Existing methods for registering these sparse images are either inaccurate or extremely slow. We present a learning-based registration method, SparseVM, that is more accurate and orders of magnitude faster than the most accurate clinical registration methods. To our knowledge, it is the first method to use deep learning specifically tailored to registering clinical images. We demonstrate our method on a clinically-acquired MRI dataset of stroke patients and on a simulated sparse MRI dataset. Our code is available as part of the VoxelMorph package at http://voxelmorph.mit.edu.","PeriodicalId":87342,"journal":{"name":"Proceedings of the ACM Conference on Health, Inference, and Learning","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2018-12-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85569998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xin Liu, Ziheng Jiang, Josh Fromm, Xuhai Xu, Shwetak N. Patel, Daniel J. McDuff
{"title":"MetaPhys","authors":"Xin Liu, Ziheng Jiang, Josh Fromm, Xuhai Xu, Shwetak N. Patel, Daniel J. McDuff","doi":"10.1145/3450439.3451870","DOIUrl":"https://doi.org/10.1145/3450439.3451870","url":null,"abstract":"","PeriodicalId":87342,"journal":{"name":"Proceedings of the ACM Conference on Health, Inference, and Learning","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77315271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Saahil Jain, Akshay Smit, S. Truong, C. Nguyen, Minh-Thanh Huynh, Mudit Jain, Victoria A Young, A. Ng, M. Lungren, P. Rajpurkar
{"title":"VisualCheXbert","authors":"Saahil Jain, Akshay Smit, S. Truong, C. Nguyen, Minh-Thanh Huynh, Mudit Jain, Victoria A Young, A. Ng, M. Lungren, P. Rajpurkar","doi":"10.1145/3450439.3451862","DOIUrl":"https://doi.org/10.1145/3450439.3451862","url":null,"abstract":"","PeriodicalId":87342,"journal":{"name":"Proceedings of the ACM Conference on Health, Inference, and Learning","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73809452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：