With the increasing application of versatile robotics, the need for end-users to teach robotic tasks via visual/video demonstrations in different environments is increasing fast. One possible method is meta-learning. However, most meta-learning methods are tailored for image classification or just focus on teaching the robot what to do, resulting in a limited ability of the robot to adapt to the real world. Thus, we propose a novel yet efficient model-agnostic meta-learning framework based on task-contrastive learning to teach the robot what to do and what not to do through positive and negative demonstrations. Our approach divides the learning procedure from visual/video demonstrations into three parts. The first part distinguishes between positive and negative demonstrations via task-contrastive learning. The second part emphasizes what the positive demo is doing, and the last part predicts what the robot needs to do. Finally, we demonstrate the effectiveness of our meta-learning approach on 1) two standard public simulated benchmarks and 2) real-world placing experiments using a UR5 robot arm, significantly outperforming current related state-of-the-art methods.

This paper proposes a novel approach to the exact symbolic analysis of very large analog circuits. The new method is based on determinant decision diagrams (DDDs) representing symbolic product terms. But instead of constructing DDD graphs directly from a flat circuit matrix, the new method constructs DDD graphs in a hierarchical way based on hierarchically defined circuit structures. The resulting algorithm can analyze much larger analog circuits exactly than before. The authors show that exact symbolic expressions of a circuit are cancellation-free expressions when the circuit is analyzed hierarchically. With this, the authors propose a novel symbolic decancellation process, which essentially leads to the hierarchical DDD graph constructions. The new algorithm partially avoids the exponential DDD construction time by employing more efficient DDD graph operations during the hierarchical construction. The experimental results show that very large analog circuits, which cannot be analyzed exactly before like /spl mu/A725 and other unstructured circuits up to 100 nodes, can be analyzed by the new approach for the first time. The new approach significantly improves the exact symbolic capacity and promises huge potentials for the applications of exact symbolic analysis.

This paper provides a novel approach to exact symbolic analysis of very large analog circuits. The new method is based on determinant decision diagrams (DDDs) to represent symbolic product terms. But instead of constructing DDD graphs directly from a flat circuit matrix, the new method constructs DDD graphs in a hierarchical way based on hierarchically defined circuit structures. The resulting algorithm can analyze much larger analog circuits exactly than before. Theoretically, we show that exact symbolic expressions of a circuit are cancellation-free expressions when the circuit is analyzed hierarchically. Practically we propose a novel hierarchical DDD graph construction algorithm. Our experimental results show that very large analog circuits, which can't be analyzed exactly before like μA725 and other unstructured circuits up to 100 nodes, can be analyzed by the new approach for the first time. The new approach significantly improves the exact symbolic capacity and promises huge potentials for the new applications of symbolic analysis in analog circuit design automation.