Stochastic Sequential Decision Making over Expanding Networks with Graph Filtering
This addresses the limitation of existing graph filtering methods that assume fixed graphs, which is incremental for domains like recommendation systems and epidemic forecasting where networks expand dynamically.
The paper tackles the problem of graph filtering on expanding networks where nodes continually attach with unknown patterns, proposing a stochastic sequential decision-making framework that adapts filtering to graph evolution and achieves benefits over batch and online alternatives in applications like cold-start recommendation and COVID prediction.
Graph filters leverage topological information to process networked data with existing methods mainly studying fixed graphs, ignoring that graphs often expand as nodes continually attach with an unknown pattern. The latter requires developing filter-based decision-making paradigms that take evolution and uncertainty into account. Existing approaches rely on either pre-designed filters or online learning, limited to a myopic view considering only past or present information. To account for future impacts, we propose a stochastic sequential decision-making framework for filtering networked data with a policy that adapts filtering to expanding graphs. By representing filter shifts as agents, we model the filter as a multi-agent system and train the policy following multi-agent reinforcement learning. This accounts for long-term rewards and captures expansion dynamics through sequential decision-making. Moreover, we develop a context-aware graph neural network to parameterize the policy, which tunes filter parameters based on information of both the graph and agents. Experiments on synthetic and real datasets from cold-start recommendation to COVID prediction highlight the benefits of using a sequential decision-making perspective over batch and online filtering alternatives.