In real-world applications, data do not reflect the ones commonly used for neural networks training, since they are usually few, unlabeled and can be available as a stream. Hence many existing deep learning solutions suffer from a limited range of applications, in particular in the case of online streaming data that evolve over time. To narrow this gap, in this work we introduce a novel and complex setting involving unsupervised meta-continual learning with unbalanced tasks. These tasks are built through a clustering procedure applied to a fitted embedding space. We exploit a meta-learning scheme that simultaneously alleviates catastrophic forgetting and favors the generalization to new tasks. Moreover, to encourage feature reuse during the meta-optimization, we exploit a single inner loop taking advantage of an aggregated representation achieved through the use of a self-attention mechanism. Experimental results on few-shot learning benchmarks show competitive performance even compared to the supervised case. Additionally, we empirically observe that in an unsupervised scenario, the small tasks and the variability in the clusters pooling play a crucial role in the generalization capability of the network. Further, on complex datasets, the exploitation of more clusters than the true number of classes leads to higher results, even compared to the ones obtained with full supervision, suggesting that a predefined partitioning into classes can miss relevant structural information.

--In this paper, we propose a deep reinforcement learning (DRL) solution to the grasping problem using 2.5D images as the only source of information. In particular, we developed a simulated environment where a robot equipped with a vacuum gripper has the aim of reaching blocks with planar surfaces. These blocks can have different dimensions, shapes, position and orientation. The experiments demonstrated the effectiveness of the proposed DRL algorithm applied to grasp tasks guided by visual depth camera inputs. When using the proper policy, the proposed method estimates a robot tool configuration that reaches the object surface with negligible position and orientation errors. This is, to the best of our knowledge, the first successful attempt of using 2.5D images only as of the input of a DRL algorithm, to solve the grasping problem regressing 3D world coordinates. I. INTRODUCTION In industrial environments, manipulator robots are usually designed to solve precise and predefined tasks. However, there are situations where it may be required to generalize the behaviour of the robots due to variations of size, shape, position, and orientation of the object to grasp. In these cases, the development of solutions according to mainstream standard computer vision and robotic control approaches can be complex and may lead to customized algorithms that cannot be easily generalized to different scenarios. Deep Reinforcement Learning addresses this task by merging the reinforcement learning and the deep learning domains, approximating the policy to learn with a deep neural network.