Recently, novelty detection with reconstruction along projection pathway (RaPP) has made progress toward leveraging hidden activation values. RaPP compares the input and its autoencoder reconstruction in hidden spaces to detect novelty samples. Nevertheless, traditional autoencoders have not yet begun to fully exploit this method. In this paper, we propose a new model, the Extended Autoencoder Model, that adds an adversarial component to the autoencoder to take full advantage of RaPP. The adversarial component matches the latent variables of the reconstructed input to the latent variables of the original input to detect novelty samples with high hidden reconstruction errors. The proposed model can be combined with variants of the autoencoder, such as a variational autoencoder or adversarial autoencoder. The effectiveness of the proposed model was evaluated across various novelty detection datasets. Our results demonstrated that extended autoencoders are capable of outperforming conventional autoencoders in detecting novelties using the RaPP method.

No abstract

In ubiquitous environment intelligent and transparent services are provided to the users anytime and anywhere. Ubiquitous systems can be effectively implemented using the agent technology, while the agent platform supports efficient and stable interaction among the agents. In this paper we propose a load-balancing scheme which considers agent state for fair and efficient resource allocation to the agents. It is achieved by the migration of agents decided according to the condition of message load and resource allocation. Experiment with an actual multi-agent system shows that the mean round trip time is significantly reduced compared to the existing scheme especially when the agents frequently change the state and transmit the messages.

No abstract

No abstract