- Rating
- 1
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 3
- Soundness
- 1
Read review
Paper summary
This paper introduces SGNN, a stochastic graph neural net for learning discrete processes such as epidemic spreading on graphs. They focus on the SI (susceptible/infected) model. The key idea is to use the reparametrization trick and Gumbel-softmax to get differentiable probability for the S/I states. The rest of the model is similar to MPNN. They use the model recurrently to simulate epidemic spreading trajectories. They want to learn the parameters of the spreading process, including the probability of infection via each edge. Their loss is basically the mean difference over time and states S/I between the actual marginal probability of node being in state at time , and the average of the simulation probabilities with model parameters . They test their model on simulated SI processes on a few synthetic and real graphs.
Critical strengths
- Using Gumbel-softmax for discrete processes on graphs makes sense.
- Their method is a reasonable first step, but with important shortcomings (see below).
- The small SI experiments are informative.
Critical weaknesses
- No comparison with any baseline methods. I'd expect even GNN or MPNN to perform reasonably well SI and get the average probabilities right.
- SI is a very basic spreading model. In particular, since , it has basically one variable. The mean-field version (not on graphs) is fully integrable and becomes logistic growth. The graph version is a tad more involved, but still very simple. More complex processes like SIR, or SEIR would be the real challenge.
- I expect the average over simulations to have a critical flaw: its value depends heavily on patient zero location in modular graphs. You can see this the Sandia and coauthorship datasets. I believe your very high error (>10%) is because these real graphs have clusters and are modular, unlike the ER, BA and WS synthetic graphs. The clusters introduce bottlenecks in the spreading process. Therefore, which cluster patient zero belongs to becomes critical in determining at finite times. This translates to large variability among simulations and the mean becomes uninformative.
- Key related work on GNN models for discrete processes on graphs is not discussed: Neural Relational Inference (Kipf et al., 2018), GTS ("DISCRETE GRAPH STRUCTURE LEARNING FOR FORE-CASTING MULTIPLE TIME SERIES" Shang et al., 2021), NodeFormer (Wu et al., 2022)...
Questions for authors
- How does your method compare to other methods I mentioned above?
- Did you run vanilla MPNN to see if it works on this data?