Reinforcement learning (RL) is mechanized to learn from experience. It solves the problem in sequential decisions by optimizing reward-punishment through experimentation of the distinct actions in an environment. Unlike supervised learning models, RL lacks static input-output mappings and the objective of minimization of a vector error. However, to find out an optimal strategy, it is crucial to learn both continuous feedback from training data and the offline rules of the experiences with no explicit dependence on online samples. In this paper, we present a study of a multi-agent RL framework which involves a Critic in semi-offline mode criticizing over an online Actor-Critic network, namely, Critic-over-Actor-Critic (CoAC) model, in finding optimal treatment plan of ICU patients as well as optimal strategy in a combative battle game. For further validation, we also examine the model in the adversarial assignment.
more »
« less
Deconfounding Actor-Critic Network with Policy Adaptation for Dynamic Treatment Regimes
Despite intense efforts in basic and clinical research, an individualized ventilation strategy for critically ill patients remains a major challenge. Recently, dynamic treatment regime (DTR) with reinforcement learning (RL) on electronic health records (EHR) has attracted interest from both the healthcare industry and machine learning research community. However, most learned DTR policies might be biased due to the existence of confounders. Although some treatment actions non-survivors received may be helpful, if confounders cause the mortality, the training of RL models guided by long-term outcomes (e.g., 90-day mortality) would punish those treatment actions causing the learned DTR policies to be suboptimal. In this study, we develop a new deconfounding actor-critic network (DAC) to learn optimal DTR policies for patients. To alleviate confounding issues, we incorporate a patient resampling module and a confounding balance module into our actor-critic framework. To avoid punishing the effective treatment actions non-survivors received, we design a short-term reward to capture patients' immediate health state changes. Combining short-term with long-term rewards could further improve the model performance. Moreover, we introduce a policy adaptation method to successfully transfer the learned model to new-source small-scale datasets. The experimental results on one semi-synthetic and two different real-world datasets show the proposed model outperforms the state-of-the-art models. The proposed model provides individualized treatment decisions for mechanical ventilation that could improve patient outcomes.
more »
« less
- Award ID(s):
- 2145625
- NSF-PAR ID:
- 10404327
- Date Published:
- Journal Name:
- KDD '22: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining
- Page Range / eLocation ID:
- 2316 - 2326
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
Abstract The wide‐scale adoption of electronic health records (EHRs) provides extensive information to support precision medicine and personalized health care. In addition to structured EHRs, we leverage free‐text clinical information extraction (IE) techniques to estimate optimal dynamic treatment regimes (DTRs), a sequence of decision rules that dictate how to individualize treatments to patients based on treatment and covariate history. The proposed IE of patient characteristics closely resembles “The clinical Text Analysis and Knowledge Extraction System” and employs named entity recognition, boundary detection, and negation annotation. It also utilizes regular expressions to extract numerical information. Combining the proposed IE with optimal DTR estimation, we extract derived patient characteristics and use tree‐based reinforcement learning (T‐RL) to estimate multistage optimal DTRs. IE significantly improved the estimation in counterfactual outcome models compared to using structured EHR data alone, which often include incomplete data, data entry errors, and other potentially unobserved risk factors. Moreover, including IE in optimal DTR estimation provides larger study cohorts and a broader pool of candidate tailoring variables. We demonstrate the performance of our proposed method via simulations and an application using clinical records to guide blood pressure control treatments among critically ill patients with severe acute hypertension. This joint estimation approach improves the accuracy of identifying the optimal treatment sequence by 14–24% compared to traditional inference without using IE, based on our simulations over various scenarios. In the blood pressure control application, we successfully extracted significant blood pressure predictors that are unobserved or partially missing from structured EHR.
-
Existing offline reinforcement learning (RL) methods face a few major challenges, particularly the distributional shift between the learned policy and the behavior policy. Offline Meta-RL is emerging as a promising approach to address these challenges, aiming to learn an informative meta-policy from a collection of tasks. Nevertheless, as shown in our empirical studies, offline Meta-RL could be outperformed by offline single-task RL methods on tasks with good quality of datasets, indicating that a right balance has to be delicately calibrated between "exploring" the out-of-distribution state-actions by following the meta-policy and "exploiting" the offline dataset by staying close to the behavior policy. Motivated by such empirical analysis, we propose model-based offline ta-RL with regularized policy optimization (MerPO), which learns a meta-model for efficient task structure inference and an informative meta-policy for safe exploration of out-of-distribution state-actions. In particular, we devise a new meta-Regularized model-based Actor-Critic (RAC) method for within-task policy optimization, as a key building block of MerPO, using both conservative policy evaluation and regularized policy improvement; and the intrinsic tradeoff therein is achieved via striking the right balance between two regularizers, one based on the behavior policy and the other on the meta-policy. We theoretically show that the learnt policy offers guaranteed improvement over both the behavior policy and the meta-policy, thus ensuring the performance improvement on new tasks via offline Meta-RL. Our experiments corroborate the superior performance of MerPO over existing offline Meta-RL methods.more » « less
-
Age-related macular degeneration (AMD) is the leading cause of irreversible blindness in developed countries. Identifying patients at high risk of progression to late AMD, the sight-threatening stage, is critical for clinical actions, including medical interventions and timely monitoring. Recently, deep-learning-based models have been developed and achieved superior performance for late AMD pre- diction. However, most existing methods are limited to the color fundus photography (CFP) from the last ophthalmic visit and do not include the longitudinal CFP history and AMD progression during the previous years’ visits. Patients in different AMD subphenotypes might have various speeds of progression in different stages of AMD disease. Capturing the progression information during the previous years’ visits might be useful for the prediction of AMD pro- gression. In this work, we propose a Contrastive-Attention-based Time-aware Long Short-Term Memory network (CAT-LSTM) to predict AMD progression. First, we adopt a convolutional neural network (CNN) model with a contrastive attention module (CA) to extract abnormal features from CFPs. Then we utilize a time-aware LSTM (T-LSTM) to model the patients’ history and consider the AMD progression information. The combination of disease pro- gression, genotype information, demographics, and CFP features are sent to T-LSTM. Moreover, we leverage an auto-encoder to represent temporal CFP sequences as fixed-size vectors and adopt k-means to cluster them into subphenotypes. We evaluate the pro- posed model based on real-world datasets, and the results show that the proposed model could achieve 0.925 on area under the receiver operating characteristic (AUROC) for 5-year late-AMD prediction and outperforms the state-of-the-art methods by more than 3%, which demonstrates the effectiveness of the proposed CAT-LSTM. After analyzing patient representation learned by an auto-encoder, we identify 3 novel subphenotypes of AMD patients with different characteristics and progression rates to late AMD, paving the way for improved personalization of AMD management. The code of CAT-LSTM can be found at GitHub .more » « less
-
The adaptive bitrate selection (ABR) mechanism, which decides the bitrate for each video chunk is an important part of video streaming. There has been significant interest in developing Reinforcement-Learning (RL) based ABR algorithms because of their ability to learn efficient bitrate actions based on past data and their demonstrated improvements over wired, 3G and 4G networks. However, the Quality of Experience (QoE), especially video stall time, of state-of-the-art ABR algorithms including the RL-based approaches falls short of expectations over commercial mmWave 5G networks, due to widely and wildly fluctuating throughput. These algorithms find optimal policies for a multi-objective unconstrained problem where the policies inherently depend on the predefined weight parameters of the multiple objectives (e.g., bitrate maximization, stall-time minimization). Our empirical evaluation suggests that such a policy cannot adequately adapt to the high variations of 5G throughput, resulting in long stall times. To address these issues, we formulate the ABR selection problem as a constrained Markov Decision Process where the objective is to maximize the QoE subject to a stall-time constraint. The strength of this formulation is that it helps mitigate the stall time while maintaining high bitrates. We propose COREL, a primal-dual actor-critic RL algorithm, which incorporates an additional critic network to estimate stall time compared to existing RL-based approaches and can tune the optimal dual variable or weight to guide the policy towards minimizing stall time. Our experiment results across various commercial mmWave 5G traces reveal that COREL reduces the average stall time by a factor of 4 and the 95th percentile by a factor of 2.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1145/3534678.3539413
