اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدMURAL: Meta-Learning Uncertainty-Aware Rewards for Outcome-Driven Reinforcement Learning
313
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
Exploration in reinforcement learning is a challenging problem: in the worst case, the agent must search for high-reward states that could be hidden anywhere in the state space. Can we define a more tractable class of RL problems, where the agent is provided with examples of successful outcomes? In this problem setting, the reward function can be obtained automatically by training a classifier to categorize states as successful or not. If trained properly, such a classifier can provide a well-shaped objective landscape that both promotes progress toward good states and provides a calibrated exploration bonus. In this work, we show that an uncertainty aware classifier can solve challenging reinforcement learning problems by both encouraging exploration and provided directed guidance towards positive outcomes. We propose a novel mechanism for obtaining these calibrated, uncertainty-aware classifiers based on an amortized technique for computing the normalized maximum likelihood (NML) distribution. To make this tractable, we propose a novel method for computing the NML distribution by using meta-learning. We show that the resulting algorithm has a number of intriguing connections to both count-based exploration methods and prior algorithms for learning reward functions, while also providing more effective guidance towards the goal. We demonstrate that our algorithm solves a number of challenging navigation and robotic manipulation tasks which prove difficult or impossible for prior methods.
قيم البحث
اقرأ أيضاً
Reinforcement learning can enable complex, adaptive behavior to be learned automatically for autonomous robotic platforms. However, practical deployment of reinforcement learning methods must contend with the fact that the training process itself can
be unsafe for the robot. In this paper, we consider the specific case of a mobile robot learning to navigate an a priori unknown environment while avoiding collisions. In order to learn collision avoidance, the robot must experience collisions at training time. However, high-speed collisions, even at training time, could damage the robot. A successful learning method must therefore proceed cautiously, experiencing only low-speed collisions until it gains confidence. To this end, we present an uncertainty-aware model-based learning algorithm that estimates the probability of collision together with a statistical estimate of uncertainty. By formulating an uncertainty-dependent cost function, we show that the algorithm naturally chooses to proceed cautiously in unfamiliar environments, and increases the velocity of the robot in settings where it has high confidence. Our predictive model is based on bootstrapped neural networks using dropout, allowing it to process raw sensory inputs from high-bandwidth sensors such as cameras. Our experimental evaluation demonstrates that our method effectively minimizes dangerous collisions at training time in an obstacle avoidance task for a simulated and real-world quadrotor, and a real-world RC car. Videos of the experiments can be found at https://sites.google.com/site/probcoll.
While reinforcement learning algorithms provide automated acquisition of optimal policies, practical application of such methods requires a number of design decisions, such as manually designing reward functions that not only define the task, but als
o provide sufficient shaping to accomplish it. In this paper, we discuss a new perspective on reinforcement learning, recasting it as the problem of inferring actions that achieve desired outcomes, rather than a problem of maximizing rewards. To solve the resulting outcome-directed inference problem, we establish a novel variational inference formulation that allows us to derive a well-shaped reward function which can be learned directly from environment interactions. From the corresponding variational objective, we also derive a new probabilistic Bellman backup operator reminiscent of the standard Bellman backup operator and use it to develop an off-policy algorithm to solve goal-directed tasks. We empirically demonstrate that this method eliminates the need to design reward functions and leads to effective goal-directed behaviors.
Despite recent success of deep network-based Reinforcement Learning (RL), it remains elusive to achieve human-level efficiency in learning novel tasks. While previous efforts attempt to address this challenge using meta-learning strategies, they typi
cally suffer from sampling inefficiency with on-policy RL algorithms or meta-overfitting with off-policy learning. In this work, we propose a novel meta-RL strategy to address those limitations. In particular, we decompose the meta-RL problem into three sub-tasks, task-exploration, task-inference and task-fulfillment, instantiated with two deep network agents and a task encoder. During meta-training, our method learns a task-conditioned actor network for task-fulfillment, an explorer network with a self-supervised reward shaping that encourages task-informative experiences in task-exploration, and a context-aware graph-based task encoder for task inference. We validate our approach with extensive experiments on several public benchmarks and the results show that our algorithm effectively performs exploration for task inference, improves sample efficiency during both training and testing, and mitigates the meta-overfitting problem.
Recent advances in supervised learning and reinforcement learning have provided new opportunities to apply related methodologies to automated driving. However, there are still challenges to achieve automated driving maneuvers in dynamically changing
environments. Supervised learning algorithms such as imitation learning can generalize to new environments by training on a large amount of labeled data, however, it can be often impractical or cost-prohibitive to obtain sufficient data for each new environment. Although reinforcement learning methods can mitigate this data-dependency issue by training the agent in a trial-and-error way, they still need to re-train policies from scratch when adapting to new environments. In this paper, we thus propose a meta reinforcement learning (MRL) method to improve the agents generalization capabilities to make automated lane-changing maneuvers at different traffic environments, which are formulated as different traffic congestion levels. Specifically, we train the model at light to moderate traffic densities and test it at a new heavy traffic density condition. We use both collision rate and success rate to quantify the safety and effectiveness of the proposed model. A benchmark model is developed based on a pretraining method, which uses the same network structure and training tasks as our proposed model for fair comparison. The simulation results shows that the proposed method achieves an overall success rate up to 20% higher than the benchmark model when it is generalized to the new environment of heavy traffic density. The collision rate is also reduced by up to 18% than the benchmark model. Finally, the proposed model shows more stable and efficient generalization capabilities adapting to the new environment, and it can achieve 100% successful rate and 0% collision rate with only a few steps of gradient updates.
Reinforcement learning provides a general framework for flexible decision making and control, but requires extensive data collection for each new task that an agent needs to learn. In other machine learning fields, such as natural language processing
or computer vision, pre-training on large, previously collected datasets to bootstrap learning for new tasks has emerged as a powerful paradigm to reduce data requirements when learning a new task. In this paper, we ask the following question: how can we enable similarly useful pre-training for RL agents? We propose a method for pre-training behavioral priors that can capture complex input-output relationships observed in successful trials from a wide range of previously seen tasks, and we show how this learned prior can be used for rapidly learning new tasks without impeding the RL agents ability to try out novel behaviors. We demonstrate the effectiveness of our approach in challenging robotic manipulation domains involving image observations and sparse reward functions, where our method outperforms prior works by a substantial margin.
الأسئلة المقترحة
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
