Time series generation is widely used in real-world applications such as simulation, data augmentation, and hypothesis testing. Recently, diffusion models have emerged as the de facto approach to time series generation, enabling diverse synthesis scenarios. However, the fixed standard-Gaussian diffusion prior may be ill-suited for time series data, which exhibit properties such as temporal order and fixed time points. In this paper, we propose TimeBridge, a framework that flexibly synthesizes time series data by using diffusion bridges to learn paths between a chosen prior and the data distribution. We then explore several prior designs tailored to time series synthesis. Our framework covers (i) data- and time-dependent priors for unconditional generation and (ii) scale-preserving priors for conditional generation. Experiments show that our framework with data-driven priors outperforms standard diffusion models on time series generation.

The WS-CDL (Web services choreography description language) from W3C facilitates the specification of rules to govern the ordering of message exchanges between Web services participants. The authors consider a computing environment where a mobile client interacts with a Web services provider according to a WS-CDL specification, and propose a framework through which the client can specify its preference on how conversation should take place. The presented framework allows for mobile clients to effectively cope with user and device mobility through providing a flexible means to reduce the number of exchanged messages without violating choreography requirement.

Due to the increased applications of agents, agent-oriented software becomes large and complex. To support systematic developments of such software, the agent-oriented software development methodology needs to be developed. This paper focuses on the modeling phase of agent-oriented software development. For agent-oriented software modeling, the Agent Elicitation method, Intra and Inter Agent modeling method are proposed. Agent Elicitation shows how to extract agents from classes in the real world. Modeling methods of agent's characteristics-Goal, Belief, Plan and Capability-are proposed for Intra Agent modeling. Methods of agent's mobility and communication in multiagent systems are proposed for Inter Agent modeling.