We aim at a robot capable to learn sequences of actions to achieve a field of complex tasks. In this paper, we are considering the learning of a set of interrelated complex tasks hierarchically organized. To learn this high-dimensional mapping between a continuous high-dimensional space of tasks and an infinite dimensional space of unbounded sequences of actions, we introduce a new framework called “procedures”, which enables the autonomous discovery of how to combine previously learned skills in order to learn increasingly complex combinations of motor policies. We propose an active learning algorithmic architecture, capable of organizing its learning process in order to achieve a field of complex tasks by learning sequences of primitive motor policies. Based on heuristics of active imitation learning, goal-babbling and strategic learning using intrinsic motivation, our algorithmic architecture leverages our procedures framework to actively decide during its learning process which outcome to focus on and which exploration strategy to apply. We show on a simulated environment that our new architecture is capable of tackling the learning of complex motor policies by adapting the complexity of its policies to the task at hand. We also show that our “procedures” enable the learning agent to discover the task hierarchy and exploit his experience of previously learned skills to learn new complex tasks.

We propose an active learning architecture for robots, capable of organizing its learning process to achieve a field of complex tasks by learning sequences of motor policies, called Intrinsically Motivated Procedure Babbling (IM-PB). The learner can generalize over its experience to continuously learn new tasks. It chooses actively what and how to learn based by empirical measures of its own progress. In this paper, we are considering the learning of a set of interrelated tasks outcomes hierarchically organized.We introduce a framework called "procedures", which are sequences of policies defined by the combination of previously learned skills . Our algorithmic architecture uses the procedures to autonomously discover how to combine simple skills to achieve complex goals. It actively chooses between 2 strategies of goaldirected exploration: exploration of the policy space or the procedural space. We show on a simulated environment that our new architecture is capable of tackling the learning of complex motor policies, to adapt the complexity of its policies to the task at hand. We also show that our "procedures" framework helps the learner to tackle difficult hierarchical tasks.

No abstract

We aim for a robot capable to learn sequences of motor policies to achieve a field of complex tasks. In this paper, we consider a set of interrelated complex tasks hierarchically organized. To address this high-dimensional mapping between a continuous high-dimensional space of tasks and an infinite dimensional space of sequences of policies, we propose a framework called "procedure", which enables the creation of sequences of policies defined as the combination of previously learned skills. We also introduce an active learning algorithmic architecture, capable of organizing its learning process in order to achieve a field of complex tasks by learning sequences of primitive motor policies. Based on heuristics of goal-babbling, social guidance, strategic learning guided by intrinsic motivation, and the "procedure" framework, our algorithm can actively decide on which outcome to focus and which exploration strategy to apply. These strategies can be autonomous exploration or active imitation, requesting demonstrations to expert teachers. We show on a simulated setup using a real-world industrial robot, that our new algorithm can tackle the learning of complex motor policies and adapt this complexity to that of the task at hand. Owing to its exploration strategies, it can discover the levels of difficulty of the tasks, and learn the hierarchy between tasks so as to combine simple tasks to complete a complex task. We also show that the procedures can replace motor policies as demonstrations for hierarchical complex tasks.

In open-ended continuous environments, robots need to learn multiple parameterised control tasks in hierarchical reinforcement learning. We hypothesise that the most complex tasks can be learned more easily by transferring knowledge from simpler tasks, and faster by adapting the complexity of the actions to the task. We propose a task-oriented representation of complex actions, called procedures, to learn online task relationships and unbounded sequences of action primitives to control the different observables of the environment. Combining both goal-babbling with imitation learning, and active learning with transfer of knowledge based on intrinsic motivation, our algorithm self-organises its learning process. It chooses at any given time a task to focus on; and what, how, when and from whom to transfer knowledge. We show with a simulation and a real industrial robot arm, in cross-task and cross-learner transfer settings, that task composition is key to tackle highly complex tasks. Task decomposition is also efficiently transferred across different embodied learners and by active imitation, where the robot requests just a small amount of demonstrations and the adequate type of information. The robot learns and exploits task dependencies so as to learn tasks of every complexity.

Multi-task learning by robots poses the challenge of the domain knowledge: complexity of tasks, complexity of the actions required, relationship between tasks for transfer learning. We demonstrate that this domain knowledge can be learned to address the challenges in life-long learning. Specifically, the hierarchy between tasks of various complexities is key to infer a curriculum from simple to composite tasks. We propose a framework for robots to learn sequences of actions of unbounded complexity in order to achieve multiple control tasks of various complexity. Our hierarchical reinforcement learning framework, named SGIM-SAHT, offers a new direction of research, and tries to unify partial implementations on robot arms and mobile robots. We outline our contributions to enable robots to map multiple control tasks to sequences of actions: representations of task dependencies, an intrinsically motivated exploration to learn task hierarchies, and active imitation learning. While learning the hierarchy of tasks, it infers its curriculum by deciding which tasks to explore first, how to transfer knowledge, and when, how and whom to imitate.