Humans and animals learn much better when the examples are not randomly presented but organized in a meaningful order which illustrates gradually more concepts, and gradually more complex ones. Here, we formalize such training strategies in the context of machine learning, and call them “curriculum learning”. In the context of recent research studying the difficulty of training in the presence of non-convex training criteria (for deep deterministic and stochastic neural networks), we explore curriculum learning in various set-ups. The experiments show that significant improvements in generalization can be achieved. We hypothesize that curriculum learning has both an effect on the speed of convergence of the training process to a minimum and, in the case of non-convex criteria, on the quality of the local minima obtained: curriculum learning can be seen as a particular form of continuation method (a general strategy for global optimization of non-convex functions). 1.

Abstract. Humans and animals acquire their wide repertoire of motor skills through an incremental learning process, during which progressively more complex skills are acquired and subsequently integrated with prior abilities. The order in which the skills are learned and the progressive manner in which they are developed play an important role in developing a final skill set. Inspired by this general idea, we develop an approach for learning motor skills based on a two-level curriculum. At the high level, the curriculum specifies an order in which different skills should be learned. At the low level, the curriculum defines a process for learning within a skill. The method is used to develop an ensemble of highly dynamic integrated motor skills for a planar articulated figure capable of doing parameterized hops, flips, rolls, and acrobatic sequences. Importantly, we demonstrate that the same curriculum can be successfully applied to significant variations of the articulated figure to yield appropriately individualized motor skill sets. 1