اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدPrivacy-Constrained Policies via Mutual Information Regularized Policy Gradients
94
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
As reinforcement learning techniques are increasingly applied to real-world decision problems, attention has turned to how these algorithms use potentially sensitive information. We consider the task of training a policy that maximizes reward while minimizing disclosure of certain sensitive state variables through the actions. We give examples of how this setting covers real-world problems in privacy for sequential decision-making. We solve this problem in the policy gradients framework by introducing a regularizer based on the mutual information (MI) between the sensitive state and the actions at a given timestep. We develop a model-based stochastic gradient estimator for optimization of privacy-constrained policies. We also discuss an alternative MI regularizer that serves as an upper bound to our main MI regularizer and can be optimized in a model-free setting. We contrast previous work in differentially-private RL to our mutual-information formulation of information disclosure. Experimental results show that our training method results in policies which hide the sensitive state.
قيم البحث
اقرأ أيضاً
In this paper, a novel pattern classification approach is proposed by regularizing the classifier learning to maximize mutual information between the classification response and the true class label. We argue that, with the learned classifier, the un
certainty of the true class label of a data sample should be reduced by knowing its classification response as much as possible. The reduced uncertainty is measured by the mutual information between the classification response and the true class label. To this end, when learning a linear classifier, we propose to maximize the mutual information between classification responses and true class labels of training samples, besides minimizing the classification error and reduc- ing the classifier complexity. An objective function is constructed by modeling mutual information with entropy estimation, and it is optimized by a gradi- ent descend method in an iterative algorithm. Experiments on two real world pattern classification problems show the significant improvements achieved by maximum mutual information regularization.
The recently proposed distribution correction estimation (DICE) family of estimators has advanced the state of the art in off-policy evaluation from behavior-agnostic data. While these estimators all perform some form of stationary distribution corre
ction, they arise from different derivations and objective functions. In this paper, we unify these estimators as regularized Lagrangians of the same linear program. The unification allows us to expand the space of DICE estimators to new alternatives that demonstrate improved performance. More importantly, by analyzing the expanded space of estimators both mathematically and empirically we find that dual solutions offer greater flexibility in navigating the tradeoff between optimization stability and estimation bias, and generally provide superior estimates in practice.
Deep neural networks (DNN) have demonstrated unprecedented success for medical imaging applications. However, due to the issue of limited dataset availability and the strict legal and ethical requirements for patient privacy protection, the broad app
lications of medical imaging classification driven by DNN with large-scale training data have been largely hindered. For example, when training the DNN from one domain (e.g., with data only from one hospital), the generalization capability to another domain (e.g., data from another hospital) could be largely lacking. In this paper, we aim to tackle this problem by developing the privacy-preserving constrained domain generalization method, aiming to improve the generalization capability under the privacy-preserving condition. In particular, We propose to improve the information aggregation process on the centralized server-side with a novel gradient alignment loss, expecting that the trained model can be better generalized to the unseen but related medical images. The rationale and effectiveness of our proposed method can be explained by connecting our proposed method with the Maximum Mean Discrepancy (MMD) which has been widely adopted as the distribution distance measurement. Experimental results on two challenging medical imaging classification tasks indicate that our method can achieve better cross-domain generalization capability compared to the state-of-the-art federated learning methods.
A reinforcement learning agent that needs to pursue different goals across episodes requires a goal-conditional policy. In addition to their potential to generalize desirable behavior to unseen goals, such policies may also enable higher-level planni
ng based on subgoals. In sparse-reward environments, the capacity to exploit information about the degree to which an arbitrary goal has been achieved while another goal was intended appears crucial to enable sample efficient learning. However, reinforcement learning agents have only recently been endowed with such capacity for hindsight. In this paper, we demonstrate how hindsight can be introduced to policy gradient methods, generalizing this idea to a broad class of successful algorithms. Our experiments on a diverse selection of sparse-reward environments show that hindsight leads to a remarkable increase in sample efficiency.
We propose a new way of deriving policy gradient updates for reinforcement learning. Our technique, based on Fourier analysis, recasts integrals that arise with expected policy gradients as convolutions and turns them into multiplications. The obtain
ed analytical solutions allow us to capture the low variance benefits of EPG in a broad range of settings. For the critic, we treat trigonometric and radial basis functions, two function families with the universal approximation property. The choice of policy can be almost arbitrary, including mixtures or hybrid continuous-discrete probability distributions. Moreover, we derive a general family of sample-based estimators for stochastic policy gradients, which unifies existing results on sample-based approximation. We believe that this technique has the potential to shape the next generation of policy gradient approaches, powered by analytical results.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
