Reinforcement learning (RL) has demonstrated its superiority in solving sequential decision-making problems. However, heavy dependence on immediate reward feedback impedes the wide application of RL. On the other hand, imitation learning (IL) tackles RL without relying on environmental supervision by leveraging external demonstrations. In practice, however, collecting sufficient expert demonstrations can be prohibitively expensive, yet the quality of demonstrations typically limits the performance of the learning policy. To address a practical scenario, in this work, we propose Self-Adaptive Imitation Learning (SAIL), which, provided with a few demonstrations from a sub-optimal teacher, can perform well in RL tasks with extremely delayed rewards, where the only reward feedback is trajectory-wise ranking. SAIL bridges the advantages of IL and RL by interactively exploiting the demonstrations to catch up with the teacher and exploring the environment to yield demonstrations that surpass the teacher. Extensive empirical results show that not only does SAIL significantly improve the sample efficiency, but it also leads to higher asymptotic performance across different continuous control tasks, compared with the state-of-the-art.
more »
« less
f-GAIL: Learning f-Divergence for Generative Adversarial Imitation Learning
Imitation learning (IL) aims to learn a policy from expert demonstrations that minimizes the discrepancy between the learner and expert behaviors. Various imitation learning algorithms have been proposed with different pre-determined divergences to quantify the discrepancy. This naturally gives rise to the following question: Given a set of expert demonstrations, which divergence can recover the expert policy more accurately with higher data efficiency? In this work, we propose f-GAIL – a new generative adversarial imitation learning model – that automatically learns a discrepancy measure from the f-divergence family as well as a policy capable of producing expert-like behaviors. Compared with IL baselines with various predefined divergence measures, f-GAIL learns better policies with higher data efficiency in six physics-based control tasks.
more »
« less
- NSF-PAR ID:
- 10225172
- Date Published:
- Journal Name:
- Advances in neural information processing systems
- ISSN:
- 1049-5258
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
more » « less
We propose a structured prediction approach for robot imitation learning from demonstrations. Among various tools for robot imitation learning, supervised learning has been observed to have a prominent role. Structured prediction is a form of supervised learning that enables learning models to operate on output spaces with complex structures. Through the lens of structured prediction, we show how robots can learn to imitate trajectories belonging to not only Euclidean spaces but also Riemannian manifolds. Exploiting ideas from information theory, we propose a class of loss functions based on the f-divergence to measure the information loss between the demonstrated and reproduced probabilistic trajectories. Different types of f-divergence will result in different policies, which we call imitation modes. Furthermore, our approach enables the incorporation of spatial and temporal trajectory modulation, which is necessary for robots to be adaptive to the change in working conditions. We benchmark our algorithm against state-of-the-art methods in terms of trajectory reproduction and adaptation. The quantitative evaluation shows that our approach outperforms other algorithms regarding both accuracy and efficiency. We also report real-world experimental results on learning manifold trajectories in a polishing task with a KUKA LWR robot arm, illustrating the effectiveness of our algorithmic framework.
-
more » « less
Agent navigation has been a crucial task in today's service and automated factories. Many efforts are to set specific rules for agents in a certain scenario to regulate the agent's behaviors. However, not all situations could be in advance considered, which might lead to terrible performance in a real-world application. In this paper, we propose CrowdGAIL, a method to learn from expert behaviors as an instructing policy, can train most 'human-like' agents in navigation problems without manually setting any reward function or beforehand regulations. First, the proposed model structure is based on generative adversarial imitation learning (GAIL), which imitates how humans take actions and move toward the target to a maximum extent, and by comparison, we prove the advantage of proximal policy optimization (PPO) to trust region policy optimization, thus, GAIL-PPO is what we base. Second, we design a special Sequential DemoBuffer compatible with the inner long short-term memory structure to apply spatiotemporal instruction on the agent's next step. Third, the paper demonstrates the potential of the model with an integrated social manner in a multi-agent scenario by considering human collision avoidance as well as social comfort distance. At last, experiments on the generated dataset from CrowdNav verify how close our model would act like a human being in the trajectory aspect and also how it could guide the multi-agents by avoiding any collision. Under the same evaluation metrics, CrowdGAIL shows better results compared with classic Social-GAN.
-
When applying imitation learning techniques to fit a policy from expert demonstrations, one can take advantage of prior stability/robustness assumptions on the expert's policy and incorporate such control-theoretic prior knowledge explicitly into the learning process. In this paper, we formulate the imitation learning of linear policies as a constrained optimization problem, and present efficient methods which can be used to enforce stability and robustness constraints during the learning processes. Specifically, we show that one can guarantee the closed-loop stability and robustness by posing linear matrix inequality (LMI) constraints on the fitted policy. Then both the projected gradient descent method and the alternating direction method of multipliers (ADMM) method can be applied to solve the resultant constrained policy fitting problem. Finally, we provide numerical results to demonstrate the effectiveness of our methods in producing linear polices with various stability and robustness guarantees.more » « less
-
In robotic manipulation, acquiring samples is extremely expensive because it often requires interacting with the real world. Traditional image-level data augmentation has shown the potential to improve sample efficiency in various machine learning tasks. However, image-level data augmentation is insufficient for an imitation learning agent to learn good manipulation policies in a reasonable amount of demonstrations. We propose Simulation-augmented Equivariant Imitation Learning (SEIL), a method that combines a novel data augmentation strategy of supplementing expert trajectories with simulated transitions and an equivariant model that exploits the O(2) symmetry in robotic manipulation. Experimental evaluations demonstrate that our method can learn non-trivial manipulation tasks within ten demonstrations and outperform the baselines by a significant margin.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
