Potential-Based Advice for Stochastic Policy Learning

Xiao, Baicen; Ramasubramanian, Bhaskar; Clark, Andrew; Hajishirzi, Hannaneh; Bushnell, Linda; Poovendran, Radha

doi:10.1109/cdc40024.2019.9030194

Cited by 5 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We show that using SAM allows agents to learn policies to complete the tasks faster, and obtain higher rewards than: i) using sparse rewards alone, and ii) a state-ofthe-art reward redistribution technique. This paper extends techniques introduced in our previous work [16] for single-agent RL to the multi-agent setting.…”

Section: Introductionmentioning

confidence: 79%

“…A potential term will not have to be added to ensure an unbiased policy gradient when utilizing look-back advice. This insight follows from Proposition 3 in [16] since we consider decentralized policies.…”

Section: Shaping Advice In Multi-agent Actor-criticmentioning

confidence: 87%

“…These works had focused on the use of potential-based methods in environments with discrete action spaces. In our previous work [16], we had developed potential-based techniques to learn stochastic policies in single-agent RL with continuous states and actions. We adapt and extend methods from [16] to develop potential-based techniques for MARL in this paper.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

Shaping Advice in Deep Multi-Agent Reinforcement Learning

Xiao¹,

Ramasubramanian²,

Poovendran³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Multi-agent reinforcement learning involves multiple agents interacting with each other and a shared environment to complete tasks. When rewards provided by the environment are sparse, agents may not receive immediate feedback on the quality of actions that they take, thereby affecting learning of policies. In this paper, we propose a method called Shaping Advice in deep Multi-agent reinforcement learning (SAM) to augment the reward signal from the environment with an additional reward termed shaping advice. The shaping advice is given by a difference of potential functions at consecutive time-steps. Each potential function is a function of observations and actions of the agents. The shaping advice needs to be specified only once at the start of training, and can be easily provided by non-experts. We show through theoretical analyses and experimental validation that the shaping advice provided by SAM does not distract agents from completing tasks specified by the environment reward. Theoretically, we prove that the convergence of policy gradients and value functions when using SAM implies the convergence of these quantities in the same environment in the absence of SAM. Experimentally, we evaluate SAM on three tasks in the multi-agent Particle World environment that have sparse rewards. We observe that using SAM results in agents learning policies to complete tasks faster, and obtain higher rewards than: i) using sparse rewards alone; ii) a state-of-the-art reward redistribution method.

show abstract

Section: Introductionmentioning

confidence: 79%

Section: Shaping Advice In Multi-agent Actor-criticmentioning

confidence: 87%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Shaping Advice in Deep Multi-Agent Reinforcement Learning

Xiao¹,

Ramasubramanian²,

Poovendran³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The aforementioned works had focused on the use of potential-based methods in environments with discrete action spaces. A preliminary version of this paper [20] introduced potential-based techniques to learn stochastic policies in single-agent RL with continuous states and actions.…”

Section: Related Workmentioning

confidence: 99%

“…We show that using shaping advice allows agents to learn policies to complete the tasks faster, and obtain higher rewards than: i) using sparse rewards alone, and ii) a state-of-the-art reward redistribution technique from [19]. Compared to a preliminary version that appeared in [20], in this paper, we develop a comprehensive framework for providing shaping advice in both, single-and multi-agent RL. We provide detailed theoretical analyses and experimental evaluations for each setting.…”

Section: Introductionmentioning

confidence: 99%

Shaping Advice in Deep Reinforcement Learning

Xiao¹,

Ramasubramanian²,

Poovendran³

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Reinforcement learning involves agents interacting with an environment to complete tasks. When rewards provided by the environment are sparse, agents may not receive immediate feedback on the quality of actions that they take, thereby affecting learning of policies. In this paper, we propose methods to augment the reward signal from the environment with an additional reward termed shaping advice in both single-and multi-agent reinforcement learning. The shaping advice is specified as a difference of potential functions at consecutive time-steps. Each potential function is a function of observations and actions of the agents. The use of potential functions is underpinned by an insight that the total potential when starting from any state and returning to the same state is always equal to zero. We show through theoretical analyses and experimental validation that the shaping advice does not distract agents from completing tasks specified by the environment reward. Theoretically, we prove that the convergence of policy gradients and value functions when using shaping advice implies the convergence of these quantities in the absence of shaping advice. We design two algorithms-Shaping Advice in Single-agent reinforcement learning (SAS) and Shaping Advice in Multi-agent reinforcement learning (SAM). Shaping advice in SAS and SAM needs to be specified only once at the start of training, and can easily be provided by non-experts. Experimentally, we evaluate SAS and SAM on two tasks in single-agent environments and three tasks in multi-agent environments that have sparse rewards. We observe that using shaping advice results in agents learning policies to complete tasks faster, and obtain higher rewards than algorithms that do not use shaping advice. Code for our experiments is available at https://github.com/baicenxiao/Shaping-Advice.

show abstract

Accelerating Reinforcement Learning using EEG-based implicit human feedback

Agarwal

Gupta

et al. 2021

Neurocomputing

View full text Add to dashboard Cite

Potential-Based Advice for Stochastic Policy Learning

Cited by 5 publications

References 17 publications

Shaping Advice in Deep Multi-Agent Reinforcement Learning

Shaping Advice in Deep Multi-Agent Reinforcement Learning

Shaping Advice in Deep Reinforcement Learning

Accelerating Reinforcement Learning using EEG-based implicit human feedback

Contact Info

Product

Resources

About