In this work, we propose a two-stage multi-agent deep deterministic policy gradient (TS-MADDPG) algorithm for communication-free, multi-agent reinforcement learning (MARL) under partial states and observations. In the first stage, we train prototype actor-critic networks using only partial states at actors. In the second stage, we incorporate partial observations resulting from prototype actions as side information at actors to enhance actor-critic training. This side information is useful to infer the unobserved states and hence, can help reduce the performance gap between a network with fully observable states and a partially observable one. Using a case study of building energy control in the power distribution network, we successfully demonstrate that the proposed TS-MADDPG can greatly improve the performance of single-stage MADDPG algorithms that use partial states only. This is the first work that utilizes partial local voltage measurements as observations to improve the MARL performance for a distributed power network.
more »
« less
Improving Actor-Critic Reinforcement Learning via Hamiltonian Monte Carlo Method
The actor-critic RL is widely used in various robotic control tasks. By viewing the actor-critic RL from the perspective of variational inference (VI), the policy network is trained to obtain the approximate posterior of actions given the optimality criteria. However, in practice, the actor-critic RL may yield suboptimal policy estimates due to the amortization gap and insufficient exploration. In this work, inspired by the previous use of Hamiltonian Monte Carlo (HMC) in VI, we propose to integrate the policy network of actor-critic RL with HMC, which is termed as Hamiltonian Policy. As such we propose to evolve actions from the base policy according to HMC, and our proposed method has many benefits. First, HMC can improve the policy distribution to better approximate the posterior and hence reduce the amortization gap. Second, HMC can also guide the exploration more to the regions of action spaces with higher Q values, enhancing the exploration efficiency. Further, instead of directly applying HMC into RL, we propose a new leapfrog operator to simulate the Hamiltonian dynamics. Finally, in safe RL problems, we find that the proposed method can not only improve the achieved return, but also reduce safety constraint violations by discarding potentially unsafe actions. With comprehensive empirical experiments on continuous control baselines, including MuJoCo and PyBullet Roboschool, we show that the proposed approach is a data-efficient and easy-to-implement improvement over previous actor-critic methods.
more »
« less
- Award ID(s):
- 1837369
- NSF-PAR ID:
- 10404292
- Date Published:
- Journal Name:
- IEEE Transactions on Artificial Intelligence
- ISSN:
- 2691-4581
- Page Range / eLocation ID:
- 1 to 12
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
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
-
null (Ed.)Off-policy deep reinforcement learning (RL) has been successful in a range of challenging domains. However, standard off-policy RL algorithms can suffer from several issues, such as instability in Qlearning and balancing exploration and exploitation. To mitigate these issues, we present SUNRISE, a simple unified ensemble method, which is compatible with various off-policy RL algorithms. SUNRISE integrates two key ingredients: (a) ensemble-based weighted Bellman backups, which re-weight target Q-values based on uncertainty estimates from a Q-ensemble, and (b) an inference method that selects actions using the highest upper-confidence bounds for efficient exploration. By enforcing the diversity between agents using Bootstrap with random initialization, we show that these different ideas are largely orthogonal and can be fruitfully integrated, together further improving the performance of existing off-policy RL algorithms, such as Soft Actor-Critic and Rainbow DQN, for both continuous and discrete control tasks on both low-dimensional and high-dimensional environments.more » « less
-
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
-
null (Ed.)In this article, we propose a novel semicentralized deep deterministic policy gradient (SCDDPG) algorithm for cooperative multiagent games. Specifically, we design a two-level actor-critic structure to help the agents with interactions and cooperation in the StarCraft combat. The local actor-critic structure is established for each kind of agents with partially observable information received from the environment. Then, the global actor-critic structure is built to provide the local design an overall view of the combat based on the limited centralized information, such as the health value. These two structures work together to generate the optimal control action for each agent and to achieve better cooperation in the games. Comparing with the fully centralized methods, this design can reduce the communication burden by only sending limited information to the global level during the learning process. Furthermore, the reward functions are also designed for both local and global structures based on the agents' attributes to further improve the learning performance in the stochastic environment. The developed method has been demonstrated on several scenarios in a real-time strategy game, i.e., StarCraft. The simulation results show that the agents can effectively cooperate with their teammates and defeat the enemies in various StarCraft scenarios.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1109/TAI.2022.3215614
