To investigate coupled oscillator model and community detection algorithm, an improvement method about the phase synchronization oscillator model and an optimized method (community detection via model modification) was proposed. By using the Kuramoto oscillator model as a basis, after joining the negative coupling strength, the nodes can be divided into several different synchronized clusters. In the synchronization process, the internal nodes in the same matrix are connected closely. In this method Kuramoto coupled oscillator model are expanded. The network can realize synchronization or be partitioned into several clusters depending on its structure. If all the nodes of the network are densely connected as a whole entity, synchronization will appear. If the network consists of several groups within which the connections are dense and between which the connections are sparser, the network will be partitioned into several clusters by their phases. The networks are divided into several communities because of this clustering phenomenon. The experiments show that the method is very promising. The simulation results in a variety of community structure that the proposed algorithm is an accurate, efficient and practical method for detecting community structure in networks.

No abstract

Many important robotics problems are partially observable in the sense that a single visual or force-feedback measurement is insufficient to reconstruct the state. Standard approaches involve learning a policy over beliefs or observation-action histories. However, both of these have drawbacks; it is expensive to track the belief online, and it is hard to learn policies directly over histories. We propose a method for policy learning under partial observability called the Belief-Grounded Network (BGN) in which an auxiliary belief-reconstruction loss incentivizes a neural network to concisely summarize its input history. Since the resulting policy is a function of the history rather than the belief, it can be executed easily at runtime. We compare BGN against several baselines on classic benchmark tasks as well as three novel robotic force-feedback tasks. BGN outperforms all other tested methods and its learned policies work well when transferred onto a physical robot.

No abstract

No abstract